<?xml version="1.0" encoding="UTF-8"?> <?oxygen RNGSchema="http://wab.aksis.uib.no/cost-a32/skjema/wab.rng"  type="xml" ?> <!DOCTYPE TEI  [ <!ENTITY % Wab_entities SYSTEM 'http://wab.aksis.uib.no/cost-a32/wab_entities.txt'   > %Wab_entities;]>  <TEI xmlns="http://www.tei-c.org/ns/1.0"  xmlns:m="http://www.w3.org/1998/Math/MathML">
 
 <teiHeader>
  
  <fileDesc>
   <titleStmt>
    <title>[DRAFT!] (2009-): Wittgenstein MS 148: Ms-148.xml</title>
    <author>Ludwig Wittgenstein</author>
    <editor><persName>Alois Pichler</persName> <orgName ref="http://wab.aksis.uib.no/">Wittgenstein Archives at the University of Bergen (WAB)</orgName>
     
    </editor> 
    <funder><name>Trinity College, Cambridge; Oxford University Press, Oxford; InteLex Corporation, Charlottesville; Uni Digital (earlier "Unifob Aksis"), Bergen; GDRE+ Hyper-Learning, Paris; COST Action A32, Brussels; eContent+ DISCOVERY, Luxembourg</name></funder>
    
    
    
    <respStmt><name>Alois Pichler</name> <resp>coordination, editorial guidelines, XML-TEI markup</resp></respStmt>
    <respStmt><name>Claus Huitfeldt, Kjersti Bj&oslash;rnestad Berg, Sindre S&oslash;rensen, MLCD project</name> 
     <resp>conversion from MECS-WIT to flat XML markup: parser</resp></respStmt>
    
    <respStmt><name>Alois Pichler</name> 
     <resp>conversion from MECS-WIT to flat XML markup: handling of overlap</resp></respStmt>
    <respStmt><name>Vemund Olstad</name> <resp>stylesheets</resp></respStmt>
    <respStmt><name>Tone Merete Bruvik</name> <resp>XML-TEI validation</resp></respStmt>
    
    <respStmt><name>&Oslash;ystein Reigem</name> <resp>PHP</resp></respStmt>
    <respStmt><name>Deirdre Smith, Alois Pichler</name> <resp>structural enrichment</resp></respStmt>
    
    
    <respStmt><name>Yngve Simmenes, William Boos</name> <resp>transcription and MECS-WIT markup 1998-99</resp></respStmt>
    
    
    
   </titleStmt>
   
   <publicationStmt>
    <availability>
     <p>Copyright holders: The Master and Fellows of Trinity College, Cambridge; Oxford University Press, Oxford; University of Bergen, Bergen; Uni Research AS (earlier Unifob AS), Bergen. Released under the Creative Commons General Public License Attribution, Non-Commercial, Share-Alike version 3 (CCPL BY-NC-SA).</p>
    </availability>
    
   </publicationStmt>
   <sourceDesc><p>The text has not been proofread since the production of the Bergen Electronic Edition (2000), though some corrections have been made also in the other parts of the text. Dating with reference to Pichler 1994, p&p;109ff.  </p></sourceDesc>
   
  </fileDesc>
  <revisionDesc>
   
   <change><date/><name/>: </change>
   
  </revisionDesc>
  
 </teiHeader>
 
 
 <facsimile>
  <surface lrx="0" lry="0" n="0" xml:id="Ms-148_FC"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,FC.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_FC-v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,FC-v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_1r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,1r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_1v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,1v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_2r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,2r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_2v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,2v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_3r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,3r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_3v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,3v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_4r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,4r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_4v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,4v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_5r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,5r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_5v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,5v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_6r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,6r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_6v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,6v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_7r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,7r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_7v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,7v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_8r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,8r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_8v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,8v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_9r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,9r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_9v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,9v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_10r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,10r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_10v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,10v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_11r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,11r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_11v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,11v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_12r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,12r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_12v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,12v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_13r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,13r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_13v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,13v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_14r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,14r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_14v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,14v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_15r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,15r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_15v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,15v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_16r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,16r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_16v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,16v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_17r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,17r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_17v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,17v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_18r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,18r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_18v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,18v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_19r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,19r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_19v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,19v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_20r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,20r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_20v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,20v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_21r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,21r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_21v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,21v.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_22r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,22r.jpg"/></surface>
 <surface lrx="1024" lry="1254" n="24" xml:id="Ms-148_22v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,22v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_23r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,23r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_23v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,23v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_24r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,24r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_24v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,24v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_25r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,25r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_25v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,25v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_26r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,26r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_26v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,26v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_27r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,27r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_27v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,27v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_28r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,28r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_28v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,28v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_29r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,29r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_29v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,29v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_30r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,30r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_30v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,30v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_31r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,31r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_31v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,31v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_32r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,32r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_32v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,32v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_33r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,33r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_33v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,33v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_34r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,34r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_34v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,34v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_35r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,35r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_35v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,35v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_36r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,36r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_36v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,36v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_37r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,37r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_37v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,37v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_38r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,38r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_38v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,38v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_39r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,39r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_39v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,39v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_40r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,40r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_40v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,40v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_41r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,41r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_41v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,41v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_42r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,42r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_42v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,42v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_43r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,43r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_43v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,43v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_44r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,44r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_44v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,44v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_45r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,45r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_45v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,45v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_46r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,46r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_46v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,46v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_47r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,47r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_47v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,47v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_48r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,48r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_48v"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,48v.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_BC-r"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,BC-r.jpg"/></surface>
 <surface lrx="1024" lry="1248" n="24" xml:id="Ms-148_BC"><graphic url="http://wab.aksis.uib.no/cost-a32_fax/148/Ms-148,BC.jpg"/></surface></facsimile>
		<text>
  <body>
   
   
    



    
    <ab n="Ms-148,1r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">  
    <pb facs="Ms-148_FC" rend="recto" n="pagename_Ms-148,FC pageref_Ms-148,1"/>
    
    <pb facs="Ms-148_FC-v" rend="recto" n="pagename_Ms-148,FC-v pageref_Ms-148,2"/>
    
     <pb facs="Ms-148_1r" rend="recto" n="pagename_Ms-148,1r pageref_Ms-148,3"/><fw type="pageno" place="top right">1&p.ord;</fw>

  
  
  <s type="es" rend="indl_2">The experience of fright appears &lp;when<lb/>
  
  
  we philosophise&rp; to be an   amorphous<lb/>
  experience behind the experience of<lb/>
   starting&p.es;</s><lb rend="hl"/>  
     <emph rend="bl_1"/> </ab>
  




<ab n="Ms-148,1r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">All I want to say is that it is<lb/>
 misleading to say that the word<lb/>
  &udq.sudq;fright&udq.eudq; signifies something which goes<lb/>
  along with the <add rend="im">experience of</add> expressing <del type="dnpc"><corr type="npcn">of</corr></del> fright&p.es;</s> <lb rend="hl"/>
<emph rend="bl_1"/></ab>


<ab n="Ms-148,1r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">There is here <del type="dnpc"><corr type="npcn">as</corr></del> <add rend="i">again</add> the queer case of<lb/>
  <del type="dnpc">that there is</del> a difference between<lb/>
  what we say, when we actually<lb/>
  <del type="dnpc"/> try to see what happens, &amp.and; what<lb/>
  we say when we think about it<lb/> 
  &lp;giving over the r<corr type="trsn"><orig type="trsn1">a</orig><reg type="trsn2">e</reg></corr>ins to
  language&rp;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,1r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The &usq.susq;far away&usq.eusq; look, the dreamy voice<lb/>
 seem to be only means for conveying<lb/>
 the real inner feeling&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,1r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Therefore there must be something<lb/> 
 else&udq.eudq; means nothing unless it expresses<lb/>
 a resolution to <del type="dnpc"><corr type="npcn">exp</corr></del> use a certain<lb/>
  mode of expression&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,1r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Suppose you tried to separate the feeling<lb/>
 which musi<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>
  gives you from hearing musi<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>&p.es;</s><lb rend="hl"/>     <emph rend="bl_1"/>
 <pb facs="Ms-148_1v" rend="verso" n="pagename_Ms-148,1v pageref_Ms-148,4"/> <fw type="pageno" place="top left">2&p.ord;</fw></ab>
    
    
    <ab n="Ms-148,1v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
    
 <s type="es" rend="indl_2"><seg xml:lang="english">Say &amp.and; mean &udq.sudq;long, long
  ago&dash;&udq.eudq;,</seg> &udq.sudq;lang<lb/>
  ist es her&dash;&udq.eudq;<seg xml:lang="english">&amp.and; now <del type="dnpc"/> put instead<lb/> 
  of these words new ones with many<lb/>
  more syl<corr type="tran">l</corr>ables &amp.and; try if you can<lb/>
  put the same meaning into the words&p.es;</seg></s><lb/> 
  <seg xml:lang="english"><s type="es"><corr type="trsn"><orig type="trsn1">p</orig><reg type="trsn2"><c type="c">P</c></reg></corr>ut instead of the copula a very<lb/>
   long word say &udq.sudq;<seg type="name">Kalamazoo</seg>&udq.eudq;<corr type="tra">&p.es;</corr></s> </seg> <lb rend="hl"/><emph rend="bl_2"/></ab>


<ab n="Ms-148,1v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_0"><seg xml:lang="lat"><add rend="el">Puell<emph rend="us1">a</emph></add>, <c type="c">P</c>oet<emph rend="us1">a</emph></seg>
  &udq.sudq;&usq.susq;masculine&usq.eusq; &amp.and;
  &usq.susq;feminine&usq.eusq; <emph rend="us1">feeling</emph>&udq.eudq;
  <del type="dnpc">&udq.sudq;</del> <choice type="dsl"><orig type="alt1"><del type="d">of</del></orig>  <orig type="alt2"><add rend="i">&usq.susq;attached&usq.eusq;
  to</add></orig></choice> <emph rend="us1">a</emph>&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,1v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">Aren&app.contr;t there two &lp;or<del type="dn"><gap extent="characters_1"/></del> more&rp; ways to<lb/>
  any event I might describe&qm.eis;</s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,1v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">We say &udq.sudq;making this gesture isn&app.contr;t
  all&udq.eudq;&p.es;</s><lb/> 
 <s type="es">The first answer is&colon; <c type="c">W</c>e are talking about<lb/>
 the <emph rend="us1">experience</emph> of making
  <choice type="em"><orig type="em1"><seg type="notation" ana="p" rend="literal">th<choice type="o"><orig type="o1">e</orig><orig type="o2">is</orig></choice></seg></orig>  <orig type="em2"><choice type="dsl"><orig type="alt1">the</orig>  <orig type="alt2">this</orig></choice></orig></choice>
  gesture&p.es;</s><lb/> 
 <s type="es">Secondly&colon; it is true that different<lb/>
 experiences can be described by<lb/>
  the same gesture; but not in the<lb/>
  sense that one is the pure one<lb/>
  &amp.and;the others consist &sp.ses;&p.es;</s><lb rend="hl"/><emph rend="bl_2"/> </ab>
 

<ab n="Ms-148,1v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_2">Wie ist es wenn man einmal die<lb/>
 besondere Klangfarbe eines Tones<lb/>
  <choice type="s"><orig type="alt1">merkt</orig>  <orig type="alt2"><add rend="i">h&ouml;rt</add></orig></choice> ein andermal nur den Klang als<lb/>
  solchen&qm.eis;</s>
 <lb rend="hl"/><lb rend="hl"/>
 <pb facs="Ms-148_2r" rend="recto" n="pagename_Ms-148,2r pageref_Ms-148,5"/><fw type="pageno">3&p.ordinal;</fw></ab>
   

<ab n="Ms-148,2r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
 
 <s type="es" rend="indl_2"><add rend="iupm"><gap extent="lines_2"/></add></s> 
 <s type="es" rend="indl_2">&udq.sudq;Ich nenne diesen Eindruck
  &usq.susq;blau&usq.eusq;&udq.eudq;&p.es;</s> <lb rend="hl"/> <emph rend="bl_2"/></ab>

<ab n="Ms-148,2r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
 <s type="es" rend="indl_2">Wie kann man denn die genaue Erfah<lb rend="shyphen"/>
 rung in
  &usq.susq;<seg xml:lang="lat">Poet<emph rend="us1">a</emph></seg> <abbr type="abb">etc&p.abb;</abbr>
  beschreiben&qm.eis;&usq.eusq;</s> <emph rend="bl_2"/></ab>


<ab n="Ms-148,2r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The philosophical problem is&colon;<lb/>
 &udq.sudq;<c type="c">W</c>hat is it that
  puzzles me<del type="dnpc">&qm.eis;&udq.eudq;</del><lb/>
  <emph rend="uw1"><choice type="s"><orig type="alt1">about</orig>  <orig type="alt2"><add rend="i">in</add></orig></choice> this matter</emph>&qm.eis;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,2r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">To give names is to lab<corr type="trsn"><orig type="trsn1">le</orig> <reg type="trsn2">el</reg></corr> things;<lb/> 
 but how does one lab<corr type="trsn"><orig type="trsn1">le</orig> <reg type="trsn2">el</reg></corr> impressions&p.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,2r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-2r.bmp" ana="graphics_Zeichen; Handschrift" rend="bitmap">148001</seg> das <corr type="trsn"><orig type="trsn1">a</orig><reg type="trsn2">A</reg></corr>uge
  &amp.und; der <corr type="trsn"><orig type="trsn1">w</orig><reg type="trsn2">W</reg></corr>ald</s> <lb rend="hl"/>
 <s type="es">Das m&auml;nnliche <emph rend="us1"><seg type="notation" ana="p" rend="literal">a</seg></emph> &amp.und; das weibliche
  <emph rend="us1"><seg type="notation" ana="p" rend="literal">a</seg></emph>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>

<ab n="Ms-148,2r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Es l&auml;&szlig;t sich &uuml;ber die <choice type="s"><orig type="alt1">besondere</orig>  <orig type="alt2"><add rend="i">bestimmte</add></orig></choice> Erfah<lb rend="shyphen"/>
 rung einiges sagen &amp.und; au&szlig;erdem<lb/> 
 scheint es etwas, &amp.und; zwar das
  <corr type="trsn"><orig type="trsn1">w</orig><reg type="trsn2">W</reg></corr>esentlich<lb rend="shyphen"/>
  ste, zu geben was sich nicht beschreiben<lb/> 
  l&auml;&szlig;t&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,2r[6]et2v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Man sagt hier, da&szlig; ein bestimmter<lb/>
 Eindruck benannt wird&p.es;</s> 
 <s type="es">Und darin<lb/> liegt etwas <corr type="trsn"><orig type="trsn1">s</orig><reg type="trsn2">S</reg></corr>eltsames &amp.und;
  <corr type="trsn"><orig type="trsn1">p</orig><reg type="trsn2">P</reg></corr>roblemati<lb rend="shyphen"/>sches&p.es;</s> 
 <s type="es">Denn es ist als w&auml;re der Eindruck <lb/>
  <pb facs="Ms-148_2v" rend="verso" n="pagename_Ms-148,2v pageref_Ms-148,6"/><fw type="pageno" place="top left">4</fw>
  etwas zu <corr type="trs"><orig type="trs1">&auml;kt</orig> <reg type="trs2">&Auml;th</reg></corr>erisches um ihn zu<lb/> benennen&p.es;</s> 
 <s type="es">&lp;Den Reichtum einer Frau<lb/> heiraten&p.es;&rp;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,2v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Du sagst Du hast einem ungreifbaren<lb/> Eindruck&p.es;</s> 
 <s type="es"><add rend="im">Ich bezweifle nicht, was Du sagst</add> <corr type="trsn"><orig type="trsn1">A</orig><reg type="trsn2">a</reg></corr>ber ich
  frage ob Du damit<lb/> etwas gesagt hast&p.es;</s> 
 <s type="es"><abbr type="abb">D&p.abb;h&p.abb;</abbr> wozu hast Du<lb/> diese Worte ge&auml;u&szlig;ert, in welchem
  Spiel&p.eis;</s> <lb rend="hl"/></ab>


<ab n="Ms-148,2v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">It is as though,  <choice type="s"><orig type="alt1">if</orig>  <orig type="alt2"><add rend="i">although</add></orig></choice> you can&app.contr;t tell<lb/> me
  exactly what happens inside you,<lb/> you can nevertheless tell me<lb/> something
  general about it&p.es;</s> 
 <s type="es">By saying<lb/> <abbr type="abb">e&p.abb;g&p.abb;</abbr> that you <add rend="i">are</add> having an
  impression<lb/> which can&app.contr;t be further described&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,2v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">As it were&colon; <c type="c">T</c>here <emph rend="us1">is</emph> something further<lb/> about it,
  only you <emph rend="us1">can&app.contr;t</emph> <emph rend="us1">say</emph> it; you<lb/> can only make the
  general statement&p.es;</s><lb/> 
 <s type="es" rend="indl_5">It is this <choice type="s"><orig type="alt1">idea</orig>  <orig type="alt2"><add rend="i">form of expression</add></orig></choice> which plays hell with<lb/>
  us&p.es;</s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 

<ab n="Ms-148,2v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-2v.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148002</seg></emph></s> <emph rend="bl_2"/> 
 <pb facs="Ms-148_3r" rend="recto" n="pagename_Ms-148,3r pageref_Ms-148,7"/><fw type="pageno">5</fw></ab>
    

    
 <ab n="Ms-148,3r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm"> 

 <s type="es" rend="indl_2"><emph rend="slilm"><del type="dnpc">It</del> &udq.sudq;<c type="c">T</c>here is not only the gesture but<lb/> a
  particular feeling which I can&app.contr;t<lb/> describe&udq.eudq;&colon; instead
  of that you might<lb/> ha<choice type="o"><orig type="o1">b</orig><orig type="o2">v</orig></choice>e said&colon; &udq.sudq;<c type="k">I</c> am
  trying to point<lb/> out a feeling to you&udq.eudq;
  <choice type="dsl"><orig type="alt1"><del type="d">which</del></orig>  <orig type="alt2"><add rend="i">this</add></orig></choice> would<lb/> be a grammatical remark showing<lb/>
  how my information is meant to be<lb/> used&p.es;</emph></s> 
 <s type="es">This is almost similar as<lb/> though I said &udq.sudq;<c type="c">T</c>his I call
  &usq.susq;<seg type="notation" ana="p" rend="literal">A</seg>&usq.eusq; &amp.and;<lb/> I am pointing out a
  <emph rend="us1">colour</emph><del type="dnpc">&rp;</del> <add rend="i">to you</add> not a
  shape&udq.eudq;<corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_1"/></ab>


    <ab n="Ms-148,3r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm"> 
 <s type="es" rend="indl_2">How can we point to the colour<lb/> &amp.and; not to the shape&qm.eis;</s> 
 <s type="es">Or to the<lb/> feeling of toothache &amp.and; not to the<lb/> tooth
  <abbr type="abb">etc&p.abb;</abbr>&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">What does one call &udq.sudq;describing<lb/> a feeling to
  someone&udq.eudq;&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Never mind the shape,&dash; look at the
  <emph rend="us1">colour</emph>&em.ees;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Was there a feeling of pastness<lb/> when you said you remembered
  &sp.ses;&qm.eis;&udq.eudq;</s><lb/> 
 <s type="es">&usq.susq;<c type="k">I</c> know of none&usq.eusq;&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">How does one point to a number, draw<lb/> attention to a number, <emph rend="us1">mean</emph>
  a number&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3r[7]et3v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">How do I call a taste
  &udq.sudq;lemon<corr type="tra">&div;</corr>taste&udq.eudq;&qm.eis;</s><lb/> 
 <pb facs="Ms-148_3v" rend="verso" n="pagename_Ms-148,3v pageref_Ms-148,8"/><fw type="pageno" place="top left">6</fw>
 <s type="es">Is it by having that taste &amp.and; saying<lb/> the words&colon; &udq.sudq;I
  call the taste &sp.ses;&udq.eudq;&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">And can I give a name to any <emph rend="us1">one</emph><lb/>
  taste<corr type="tra">&div;</corr>experience <del type="dnpc">as</del> without giving<lb/> the
  taste<del type="dn">t</del> a common  langu<lb rend="shyphen"/>
  age&qm.eis;&dash;</s> 
 <s type="es">&udq.sudq;<c type="k">I</c> give my feeling a name,<lb/> nobody else can know what the<lb/>
  name means&p.es;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">A <choice type="o"><orig type="o1"><unclear/></orig> <orig type="o2">s<corr type="npcn">c</corr></orig></choice>lave has to remind me of<lb/> something
  &amp.and; isn&app.contr;t to know what<lb/> he reminds me of&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,3v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2"><emph rend="slilm"><c type="k">I</c> note down a word <add rend="im">in my diary</add> which serves<lb/> to bring back a
  taste&p.es;</emph><lb rend="hl"/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-3v.bmp" ana="graphics_Kreise; Kreis" rend="bitmap">148003</seg></emph></s> <emph rend="bl_5"/>
 <pb facs="Ms-148_4r" rend="recto" n="pagename_Ms-148,4r pageref_Ms-148,9"/><fw type="pageno"/> </ab>
 

<ab n="Ms-148,4r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">

 <s type="es"><emph rend="centered">
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-4r-a.bmp" ana="graphics_F&auml;den;  zweigeteilt" rend="bitmap">148004</seg> 
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-4r-b.bmp" ana="graphics_Vierecke;  eingerahmter Text" rend="bitmap">148005</seg></emph></s> 
 <s type="es"><emph rend="centered">
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-4r-c.bmp" ana="check_this could be transcribed I believe H  etc" rend="bitmap">k148001</seg></emph></s> 
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-4r-d.bmp" ana="graphics_Schema; mit horizontalen und vertikalen  Linien" rend="bitmap">148006</seg></emph></s> 
 <fw type="pageno"/> <pb facs="Ms-148_4v" rend="verso" n="pagename_Ms-148,4v pageref_Ms-148,10"/><fw type="pageno"/></ab>
  
 

<ab n="Ms-148,4v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">

 <s type="es" rend="indl_2"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-4v.bmp" ana="graphics_Schema; mit horizontalen und vertikalen  Linien" rend="bitmap">148007</seg></emph></s> <lb rend="hl"/>
 <pb facs="Ms-148_5r" rend="recto" n="pagename_Ms-148,5r pageref_Ms-148,11"/><fw type="pageno">7</fw></ab>

  

<ab n="Ms-148,5r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">

   <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> use the name for the impression<lb/>
    <unclear>direct<corr type="npcn">a</corr>ly</unclear> &amp.and; not in such a way<lb/> that anyone
    else can under<lb rend="shyphen"/>stand it&p.es;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
  
 

<ab n="Ms-148,5r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
  <s type="es" rend="indl_2">Buying something from oneself&p.es;</s> 
 <s type="es">Going<lb/> through the operations of buying&p.es;</s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,5r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
  <s type="es" rend="indl_2">My right hand selling to my left<lb/> hand&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>

<ab n="Ms-148,5r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2"><c type="k">G</c>ef&uuml;hls&div;&lp;Gedanken&rp;&uuml;bertragung&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>

<ab n="Ms-148,5r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Eine gute <corr type="trsn"><orig type="trsn1">a</orig><reg type="trsn2">A</reg></corr>rt eine Farbe zu<lb/> benennen w&auml;re, in einer
  entsprechend<lb/> gef&auml;rbten Tinte den Namen schreiben&p.es;</s><emph rend="bl_1"/><lb rend="hl"/> </ab>


<ab n="Ms-148,5r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> name the feeling&udq.eudq;&dash; I
  don<corr type="tra">&app.contr;</corr>t quite<lb/> know how you do this, what use you<lb/> are
  making of the <choice type="s"><orig type="alt1">word</orig>  <orig type="alt2"><add rend="i">name</add></orig></choice><corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,5r[7]" ana="date_19341000-19350500" rend="blbef_0 blaft_3" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c>&app.contr;m giving the feeling, which <choice type="s"><orig type="alt1">I
  have</orig>  <orig type="alt2"><add rend="i">I&app.contr;m having</add></orig></choice> just<lb/> now a
  name&udq.eudq;&p.es;&dash;</s> 
 <s type="es"><c type="k">I</c> don&app.contr;t quite know<lb/> what you are doing&p.es;</s> <lb rend="hl"/> <emph rend="bl_3"/></ab>



<ab n="Ms-148,5r[8]et5v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_4" xml:lang="english">
 <s type="es" rend="indl_2">One might say&colon; &udq.sudq;<c type="c">W</c>hat is the use<lb/> of talking of our
  feeling at all&p.eis;</s> 
 <s type="es">Let 
  <pb facs="Ms-148_5v" rend="verso" n="pagename_Ms-148,5v pageref_Ms-148,12"/><fw type="pageno" place="top left">8</fw> us devise a language which really
  only<lb/> says what can be understood<corr type="tra">&p.es;</corr>&udq.eudq;</s> <lb/>
 <s type="es">Thus I am not to say &udq.sudq;I have a<lb/> feeling of
  pastness&udq.eudq;&colon; <c type="c">B</c>ut</s> <lb rend="hl"/><emph rend="bl_4"/></ab>
 


<ab n="Ms-148,5v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm"><emph rend="indl_2"/>
 <s type="es">&udq.sudq;This pain I call &usq.susq;toothache&usq.eusq; &amp.and; I can<lb/>
  never make him understand what<lb/> it means&udq.eudq;&p.es;</s>  <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,5v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">We are under the impression that we can<lb/> point to the pain, as it were
  unseen by<lb/> the other person, &amp.and; name it&p.es;</s> </emph> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,5v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">For what does it mean that this <choice type="s"><orig type="alt1">pain</orig>  <orig type="alt2"><add rend="i">feeling</add></orig></choice><lb/> is the
  meaning of this name&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,5v[5]et6r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Or, that the pain is the b<corr type="tran">e</corr>arer of the<lb/> name&qm.eis;</s> <lb rend="hl"/>
 
 <s type="es" rend="indl_5">It is the substantive &usq.susq;pain&usq.eusq; which<lb/> <del type="dnpc"><gap extent="words_1"/></del>
  puzzles us&p.es;</s> 
 <s type="es">This substantive seems<lb/> to produce an i<corr type="tran">l</corr>lusion&p.es;</s> 
 <s type="es">What would<lb/> things look like if we expressed pains<lb/> by moaning &amp.and;
  holding the painful spot&qm.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Or that we utter the word pain pointing<lb/> to a spot&p.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4">&udq.sudq;But that the point is that we should<lb/> 
  <pb facs="Ms-148_6r" rend="recto" n="pagename_Ms-148,6r pageref_Ms-148,13"/><fw type="pageno">9</fw> say &usq.susq;pain&usq.eusq; when there
  really is<del type="dn"><gap extent="characters_1"/></del> pain&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">But how am <emph rend="us1">I</emph> to know if there really is<lb/> pain&qm.eis; if what I
  feel really is pain&qm.eis;</s> 
 <s type="es">Or,<lb/> if I really have a <emph rend="us1">feeling</emph>&qm.eis;&sdash;</s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 

<ab n="Ms-148,6r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">Es ist sehr n&uuml;tzlich zu bedenken&colon; <c type="c">W</c>ie w&uuml;rde<lb/> ich in einer
  Geb&auml;rdensprache ausdr&uuml;cken&colon;<lb/> &udq.sudq;ich hatte keine Schmerzen, aber
  stellte<lb/> mich, als ob ich welche hatte&udq.eudq;&qm.eis;</s> </emph><lb rend="hl"/></ab>


<ab n="Ms-148,6r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_5"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-6r.bmp" ana="graphics_Abbild; Mensch&lp;en&rp;" rend="bitmap">148008</seg>
  &udq.sudq;<c type="c">S</c><corr type="npcn">h</corr>urely it isn&app.contr;t enough that he moans;<lb/>
  I must be able to describe the<lb/> state when he moans &amp.and;
  hasn&app.contr;t got pains<corr type="tra">&p.es;</corr>&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,6r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;He has pains, says he has pains &amp.and; saying<lb/>
  &usq.susq;pains&usq.eusq; <emph rend="us1">he means his</emph> pains&p.es;&udq.eudq;</s> 
 <s type="es">How does he<lb/> <emph rend="us1">mean</emph> his pains by the word &usq.susq;pain&usq.eusq;
  or<lb/> &usq.susq;toothache&usq.eusq;&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,6r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;He sa<corr type="trsn"><orig type="trsn1">i</orig><reg type="trsn2">y</reg></corr>s &usq.susq;I see green&usq.eusq;
  &amp.and; means the colour<lb/> he sees&p.es;&udq.eudq;&dash;</s> 
 <s type="es">If asked afterwards what<lb/> did you mean by &usq.susq;green&usq.eusq; he
  might answer<lb/> &usq.susq;I meant the colour&usq.eusq;, pointing to it&p.es;</s><lb rend="hl"/>
 <emph rend="bl_1"/><pb facs="Ms-148_6v" rend="verso" n="pagename_Ms-148,6v pageref_Ms-148,14"/><fw type="pageno" place="top left">10</fw></ab>
     


<ab n="Ms-148,6v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;In my own <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>ase I know that when<lb/> I say &usq.susq;I
  have pain&usq.eusq; this utterance<lb/> is accompanied by something;&dash; but<lb/> it
  is also accompanied by something<lb/> in another man&qm.eis;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">In as much as his utterance needn&app.contr;t<lb/> be accompanied by
  <emph rend="us1">my</emph> pain&p.es;</s> 
 <s type="es"><c type="k">I</c> may<lb/> say that it isn&app.contr;t accompanied by<lb/> anything&p.es;</s> <lb rend="hl"/>
<emph rend="bl_1"/></ab>


<ab n="Ms-148,6v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> know what I mean by &usq.susq;toothache&usq.eusq;<lb/> but
  the other person can&app.contr;t <emph rend="us1">know</emph> it&p.es;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_2"/></ab>
 

<ab n="Ms-148,6v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Als Negation&colon; &udq.sudq;<seg xml:lang="english"><c type="c">T</c>he deuce he
  is&sp.ses;&p.es;&udq.eudq;</seg></s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,6v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Die Philosophie eines Stammes der<lb/> als Negation <add rend="i">nur</add> den
  Ausdruck<add rend="i"><del type="dnpc"><gap extent="words_1"/></del></add> <choice type="s"><orig type="alt1">ben&uuml;tzt</orig>  <orig type="alt2"><add rend="i">kennt</add></orig></choice>&colon;<lb/>
  &udq.sudq;<seg xml:lang="english"><c type="k">I</c>&app.contr;ll be damned
   if&sp.ses;&udq.eudq;&p.es;</seg></s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,6v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2"><seg xml:lang="english"><unclear><seg xml:lang="french">On a beau
  dire</seg>&sp.ses;&p.es;</unclear></seg></s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,6v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_2">&udq.sudq;Man kann nie einen <add rend="im">ganzen</add> K&ouml;rper sehen<lb/> sondern nur
  <add rend="im">immer</add> einen Teil seiner Ober<lb rend="shyphen"/>fl&auml;che&p.es;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_4"/>
 <pb facs="Ms-148_7r" rend="recto" n="pagename_Ms-148,7r pageref_Ms-148,15"/><fw type="pageno"/></ab>
    

<ab n="Ms-148,7r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">

 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-7r.bmp" ana="graphics_Technische Darstellung; geometrischer Beweis" rend="bitmap">k148002</seg></emph></s> <lb rend="hl"/>
 <pb facs="Ms-148_7v" rend="verso" n="pagename_Ms-148,7v pageref_Ms-148,16"/><fw type="pageno" place="top right">1&p.ordinal;</fw></ab>
   


<ab n="Ms-148,7v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <emph rend="indl_2"/><add rend="iupm">
 <s type="es">They have the same <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr> if to one 1<lb/>
  there always <emph rend="us1">corresponds</emph> one&p.es;</s><lb/> <emph rend="slilm">
 <s type="es">If for one of these there <emph rend="us1">is</emph> always<lb/> one of the others&p.es;</s> 
 </emph></add><lb rend="hl"/>
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-7v-a.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148009</seg></emph></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;Two rods are equally long if for any<lb/> inch of the one there is an
  inch<lb/> of the other&udq.eudq;&p.es;</s><lb rend="hl"/> 
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-7v-b.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148010</seg></emph></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;There are these couples, whether I<lb/> write them down or
  not&p.es;&udq.eudq;</s> <lb rend="hl"/>
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-7v-c.bmp" ana="graphics_Vierecke; horizontales Rechteck" rend="bitmap">148011</seg></emph></s> <lb rend="hl"/>
 <pb facs="Ms-148_8r" rend="recto" n="pagename_Ms-148,8r pageref_Ms-148,17"/><fw type="pageno">2&p.ordinal;</fw></ab>
  


<ab n="Ms-148,8r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;Give the impression a name&em.ees;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">that seems to have sense&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">&udq.sudq;It seems to me that I can mean<lb/> the
  impression&udq.eudq;<corr type="tra">&p.es;</corr></s> <lb rend="hl"/>
 <s type="es">It seems to me that I can will<lb/> the table to app<corr type="tran">r</corr>oach&p.es;</s> 
 <emph rend="bl_1"/> <lb rend="hl"/>
 <s type="es" rend="indl_2">&udq.sudq;Can one push air&qm.eis;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_21"/> 
 <pb facs="Ms-148_8v" rend="verso" n="pagename_Ms-148,8v pageref_Ms-148,18"/><fw type="pageno" place="top left">3&p.ordinal;</fw></ab>
    
   


<ab n="Ms-148,8v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">

 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-8v-a.bmp" ana="check_one bitmap for the whole group, or refer reader to facsimile and use bitmap containing that instruction mentioned above" rend="bitmap"> k148003</seg></emph></s> <lb rend="hl"/></ab>
 


<ab n="Ms-148,8v[2]et9r[1]et9v[1]et10r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">For each member of <seg type="notation" ana="logic_type&div;theoretic class" rend="literal">&alpha;</seg> there <emph rend="us1">is</emph> a member<lb/> of
  <seg type="notation" ana="logic_type&div;theoretic class" rend="literal">&beta;</seg>&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">The <choice type="o"><orig type="o1">K</orig><orig type="o2">c</orig></choice>lasses <seg type="notation" ana="logic_type&div;theoretic class" rend="literal">&alpha;</seg> &amp.and;
  <seg type="notation" ana="logic_type&div;theoretic class" rend="literal">&beta;</seg>
  fall into couples&p.es;</s><lb/> 
 <s type="es" rend="indl_2">This similar with a <abbr corresp="proposition">prop</abbr> of say physics,<lb/>
  <abbr type="abb">e&p.abb;g&p.abb;</abbr> &udq.sudq;they join &amp.and; form couples when
  they are<lb/> brought together&udq.eudq;&p.es;</s> <lb rend="hl"/>
 <s type="es" rend="indl_2">But this is just <emph rend="us1">not</emph> what is meant&p.es;</s> <lb/>
 <s type="es">We mean something that follows from<lb/> what there exists in these
  <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>lasses&p.es;</s> 
 <s type="es">And<lb/> we have an image of them something<lb/> like this<corr type="tra">&colon;</corr><lb rend="hl"/>
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-8v-b.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148012</seg></emph></s><lb rend="hl"/> 
 <s type="es">If <add rend="im">now</add> we say for this there is this, for this<lb/> there is this
  <abbr type="abb">etc&p.abb;</abbr>, this sounds as if we<lb/> said something about the dots;
  like<lb/> &udq.sudq;this belongs to this <abbr type="abb">etc&p.abb;</abbr>&udq.eudq; whereas
  we 
  <pb facs="Ms-148_9r" rend="recto" n="pagename_Ms-148,9r pageref_Ms-148,19"/><fw type="pageno">4</fw> are saying words &amp.and;
  gestures to put them<lb/> into couples&p.es;</s> 
 <s type="es">And this <del type="dnpc">of</del> <emph rend="us1">is</emph> a way of<lb/> finding whether they have
  the same <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr><corr type="tra">&p.es;</corr></s> <lb rend="hl"/>
 
 <s type="es" rend="indl_2">And now we must say that there <add rend="i">are many different</add> phenomena<lb/>
  <del type="dnpc">of</del> of equality of <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr> or
  of having a certain<lb/>
  <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr><corr type="tra">&p.es;</corr></s> 
 <s type="es">Just as having a length &amp.and; having<lb/> equal lengths&p.es;</s> 
 <s type="es">Let me remin<choice type="o"><orig type="o1">g</orig><orig type="o2">d</orig></choice> you<lb/> of the problem &udq.sudq;are these two
  rods<lb/> <emph rend="us1">now</emph> of the same length&p.eis;&udq.eudq;</s> 
 <s type="es">Take the definition<lb/> you have to give of this expression<lb/> when the rods have
  to be measured<lb/> &amp.and; on the other hand when you use this<lb/>
  difference&p.es;</s> 
 <s type="es">&udq.sudq;These bodies have the same<lb/> weight&udq.eudq;
  <abbr type="abb">etc&p.abb;</abbr><corr type="tra">&p.es;</corr></s> 
 <s type="es">Now consider&colon; &udq.sudq;<c type="c">T</c>here are as<lb/> many
  gra<corr type="tran">i</corr>n<corr type="npcn">e</corr>s of sand in this heap as<lb/> in the
  other&udq.eudq;&p.es;</s> 
 <s type="es">How do we know this&qm.eis;</s><lb/> 
 <s type="es">&lp;This is no psychological question&p.es;&rp;</s><lb/> 
 <s type="es">Now suppose we said we test it by<lb/> connecting the classes one&div;one;
  <unclear>then</unclear> the <lb/>question is&colon; how shall we know that<lb/> we have
  connected them&qm.eis;</s> 
 <s type="es">For there<lb/> are several utterly different criteria&p.es;</s><lb/> 
 <s type="es">But further what shall we say in the<lb/> cases where no such connection is
  possi<lb rend="shyphen"/>ble&qm.eis;</s> 
 <s type="es"><del type="dnpc"><gap extent="words_1"/></del> <c type="c">W</c>hat about saying then that<lb/> the members of the two
  similar classes<lb/> still fall into couples&qm.eis;&qm.eis;</s> 
 <s type="es">Is this now<lb/> an explanation&qm.eis;</s> 
 <s type="es">For when we give it we<lb/> thought of it as reducing the statement
  <pb facs="Ms-148_9v" rend="verso" n="pagename_Ms-148,9v pageref_Ms-148,20"/><fw type="pageno" place="top left">5</fw> of numeric equality to simpler terms&p.es;</s><lb/>
 
 <s type="es">The falling into couples was an<lb/> image which in some cases was most<lb/> natural
  namely in those in which<lb/> there was the possibility of joining<lb/> terms into
  couples&p.es;</s> 
 <s type="es">But in fact<lb/> it wasn<corr type="tra">&app.contr;</corr>t at all the only<lb/> aspect of
  numeric equality&p.es;</s> 
 <s type="es">The<lb/> term &udq.sudq;having the same
  <abbr corresp="number"><choice type="o"><orig type="o1">n</orig><orig type="o2">N</orig></choice><emph rend="sup">r</emph></abbr>&udq.eudq; in fact<lb/>
  suggests a different aspect&p.es;</s> 
 <s type="es"><c type="k">I</c> mean this<lb rend="hl"/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-9v-a.bmp" ana="graphics_Schema; ohne Linien" rend="bitmap">148013</seg></emph></s> <lb rend="hl"/>
 <s type="es">Having the same <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr> can be
  interpreted<lb/> as having the same one of these<lb/> schemata&p.es;</s> 
 <s type="es">Of course this aspect<lb/> too is only natural in a very limited
  <choice type="o"><orig type="o1"><abbr type="abb"><c type="c">N</c><emph rend="sup">r</emph>&p.abb;</abbr></orig><orig type="o2">number</orig></choice> of cases&p.es;</s> 
 <s type="es"><add rend="el">Aspect of <emph rend="us1">stars</emph></add>&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">The explanation, that two classes<lb/> have the same
  <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr> if they fall into<lb/> couples, is
  really taken from <corr type="npc">the</corr><lb/> a case like<lb rend="hl"/>
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-9v-b.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148014</seg>
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-9v-c.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148015</seg></emph></s> <lb rend="hl"/>
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-9v-d.bmp" ana="graphics_Vielecke; Stern mit f&uuml;nf Punkten" rend="bitmap">148016</seg></emph> <del type="dnpc"><c type="c">H</c>ow</del>
  &udq.sudq;<c type="c">T</c>he pentagram has twice<lb/> as many points
  <corr type="trs"><orig type="trs1">than</orig> <reg type="trs2">as</reg></corr> the  
  <pb facs="Ms-148_10r" rend="recto" n="pagename_Ms-148,10r pageref_Ms-148,21"/><fw type="pageno">6&p.ordinal;</fw>
  pentagon&udq.eudq;<corr type="tra">&p.es;</corr></s> 
 <s type="es">Demonstration</s> <lb rend="hl"/> 
 <s type="es" rend="indl_2">Timelessness of the demonstrated<lb/> <abbr corresp="proposition">prop</abbr>&colon;
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-10r.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148017</seg></emph><lb rend="hl"/>
  <s type="es">blue &amp.and; red &equ; purple<corr type="tra">&p.es;</corr></s> </s> <lb rend="hl"/>
 <pb facs="Ms-148_10v" rend="verso" n="pagename_Ms-148,10v pageref_Ms-148,22"/><fw type="pageno" place="top left">7</fw></ab>
    


<ab n="Ms-148,10v[1]et11r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_2"><unclear>We said</unclear> that what we described<lb/> as &udq.sudq;numeral
  equality&udq.eudq;, &udq.sudq;being<lb/> 1&div;1 correlated&udq.eudq;,
  &udq.sudq;having the number<lb/> <seg type="notation" ana="p" rend="literal">n</seg>&udq.eudq; were widely differing
  phenomena&p.es;</s><lb/> 
 <s type="es">That therefore it was an i<corr type="tran">l</corr>lusion<lb/> to think that to say
  &udq.sudq;the classes<lb/> fall in pairs&udq.eudq; is, generally speaking<lb/> an
  analysis of what we call<lb/> numeric equality in simpler terms&p.es;</s><lb/> 
 <s type="es">We can if we like put &udq.sudq;being numerically<lb/> equal&udq.eudq; &equ;
  &udq.sudq;falling into pairs&udq.eudq; but<lb/> the <add rend="ib">use of the</add> one
  expression just as of the<lb/> other has got to be explained<lb/> in the particular
  case&p.es;</s> 
 <s type="es">This we<lb/> only forge<corr type="trs"><orig type="trs1">tt</orig> <reg type="trs2">t</reg></corr>&p.es;</s> 
 <s type="es">Thinking about a very<lb/> special class of examples&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-10v.bmp" ana="graphics_Schema; ohne Linien" rend="bitmap">148018</seg></emph> <c type="c">T</c>he ideas that if
  they<lb/> ha<choice type="o"><orig type="o1">b</orig><orig type="o2">v</orig></choice>e the right existential<lb/> structure they do fall<lb/> into
  cou<choice type="o"><orig type="o1">l</orig><orig type="o2">p</orig></choice>les &amp.and; <emph rend="us1">this</emph> is<lb/>
  <emph rend="us1">demonstrated</emph>&p.es;</s> 
 <s type="es">The<lb/> question how we find out in the<lb/> special case that they do have<lb/> the
  right s<choice type="o"><orig type="o1">p</orig><orig type="o2">t</orig></choice>ructures is neglected&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">One could also say that a<lb/> length <add rend="i"><seg type="notation" ana="p" rend="literal">a</seg></add> was twice another one
  <seg type="notation" ana="p" rend="literal">b</seg><lb/> if <del type="dnpc">the</del> two <add rend="i"><seg type="notation" ana="p" rend="literal">a</seg></add> superimposed gave<lb/>
  <seg type="notation" ana="p" rend="literal">b</seg>&p.es;</s> 
 <s type="es">Application for wavelengths&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_3">This brings me to the topic of 
  <pb facs="Ms-148_11r" rend="recto" n="pagename_Ms-148,11r pageref_Ms-148,23"/><fw type="pageno">8</fw> demonstration&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_5">1&rp; <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11r-a.bmp" ana="graphics_Vielecke; Stern mit f&uuml;nf Punkten" rend="bitmap">148019</seg></emph>
  <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr> of outer
  <abbr corresp="vertices">vert<corr type="tra">&p.abb;</corr></abbr> &equ;
  <del type="dnpc"><abbr corresp="number">N<emph rend="sup">r</emph></abbr> of<lb/> inner
  <abbr corresp="vertices">vert<corr type="tra">&p.abb;</corr></abbr></del> 5&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/>
 
 <s type="es" rend="indl_3">Compare with &udq.sudq;the <c type="c">H</c>and has 5
  <c type="c">F</c>ingers&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Timelessness&p.es;</s> 
 <s type="es">The sa<corr type="trs"><orig type="trs1">v</orig> <reg type="trs2">m</reg></corr>e hold<corr type="tran">s</corr> of<lb/> &udq.sudq;<c type="c">T</c>he number
  of outer <abbr corresp="vertices">vert&p.abb;</abbr> &equ;
  <abbr corresp="number">N<emph rend="sup">r</emph></abbr> of inner<lb/>
  <abbr corresp="vertices">vert</abbr>&udq.eudq;<corr type="tra">&p.es;</corr></s> <lb rend="hl"/>
 <s type="es" rend="indl_4">Question which is answered by<lb/> this <abbr corresp="proposition">prop</abbr>
  timeless&p.es;</s> 
 <s type="es"><add rend="im">Apparent</add> <corr type="trsn"><orig type="trsn1">G</orig><reg type="trsn2">g</reg></corr>enerality of
  <abbr corresp="demonstration">demonstr&p.abb;</abbr><corr type="tra">&p.es;</corr></s> <lb rend="hl"/>
 
 <s type="es" rend="indl_4">The copula has no tenses&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2"><gap extent="words_1"/> idea is that the idea of a penta<lb rend="shyphen"/>gram is bound up with a
  cardinal<lb/> <abbr corresp="number">N<emph rend="sup">r</emph>&p.abb;</abbr><corr type="tra">&p.es;</corr></s> 
 <s type="es">Now, we could make all sorts of<lb/> connections&p.es;</s> <lb rend="hl"/>
 <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11r-b.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148020</seg></emph>
 <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11r-c.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148021</seg></emph><lb rend="hl"/>
 <s type="es">&udq.sudq;It is the essence of these figures<lb/> to be capable of being
  <choice type="s"><orig type="alt1">d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided</orig>  <orig type="alt2"><add rend="i">connected</add></orig></choice> in<lb/> this
  way&udq.eudq;&p.es;</s> <lb rend="hl"/><emph rend="bl_3"/>
 <pb facs="Ms-148_11v" rend="verso" n="pagename_Ms-148,11v pageref_Ms-148,24"/><fw type="pageno" place="top left">9</fw></ab>
    


<ab n="Ms-148,11v[1]et12r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 
 <s type="es" rend="indl_2"><emph rend="slilm"><emph rend="us2"><persName key="Pythagoras" corresp="commentary">Pythagoras</persName></emph></emph></s><lb rend="hl"/> <emph rend="bl_1"/>
 
 <s type="es" rend="indl_2">Is the result of the process taken<lb/> as a standard or not&p.eis;</s> <lb rend="hl"/>
 <emph rend="bl_1"/> 
 <s type="es" rend="indl_5"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11v-a.bmp" ana="graphics_Vierecke; vertikales Rechteck" rend="bitmap">148022</seg></emph> &udq.sudq;<c type="c">T</c>hese two
  triangles by <unclear>this</unclear><lb/> nature <emph rend="us1">give</emph> the
  rectangle&udq.eudq;&p.es;</s><lb rend="hl"/> 
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11v-b.bmp" ana="check_group of images on one bitmap, or use bitmap to direct the reader to the facsimile" rend="bitmap">m148001</seg></emph><lb rend="hl"/>
  <s type="es">This aspect might never have struck<lb/> you&p.es;</s><lb rend="hl"/> 
  <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-11v-c.bmp" ana="check_group of images on one bitmap, or use bitmap  to direct the reader to the facsimile" rend="bitmap">m148002</seg></s> 
  <pb facs="Ms-148_12r" rend="recto" n="pagename_Ms-148,12r pageref_Ms-148,25"/><fw type="pageno">10</fw>
  <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12r-a.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148023</seg>
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12r-b.bmp" ana="graphics_Technische Darstellung; geometrischer  Beweis" rend="bitmap">148024</seg>
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12r-c.bmp" ana="graphics_Vierecke; vertikales  Rechteck" rend="bitmap">148025</seg> <seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a
   &plus; &lp;<gap extent="words_1"/>&rp; &equ; &lp;<gap extent="words_1"/>&plus; b&rp; &plus;c</seg></seg><lb rend="hl"/>
   <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;1&rp; &equ; &lp;a&plus;b&rp;
   &plus;1</seg></seg><lb rend="hl"/> <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a &plus; &lp;b&plus;2&rp; &equ;
   &lp;a&plus;b&rp;&plus;2</seg></seg></s> <lb rend="hl"/> 
  <s type="es" rend="indl_2">It <emph rend="us1">seems</emph> you can&app.contr;t get out&p.es;</s> 
  <s type="es">You<lb/> <emph rend="us1">must</emph> adopt <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal"> <seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;c&rp; &equ;
   &lp;a&plus;b&rp;&plus;c</seg></seg> if you<lb/> adopt
   <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;1&rp; &equ; &lp;a&plus;b&rp;
   &plus;1</seg></seg><corr type="tra">&p.es;</corr></s> </s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,12r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">But <emph rend="us1">need</emph> we really say that<lb/> <seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;2&rp; &equ;
  &lp;a&plus;b&rp;&plus;2</seg></seg> follows from<lb/>
  <seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;1&rp; &equ;
  &lp;a&plus;b&rp;&plus;1</seg></seg>&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,12r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">The reasoning is&colon;<lb rend="hl"/> <emph rend="indl_5"/> <seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;&lp;b&plus;2&rp; &equ;
  a&plus;&lp;b&plus;&lp;1&plus;1&rp;&rp; &equ; a&plus; &lp;&lp;b&plus;1&rp;
  &plus;1&rp; &equ;<lb rend="hl"/> &equ; &lp;a&plus; &lp;b&plus;1&rp;&rp; &plus;1 &equ;
  &lp;&lp;a&plus;b&rp; &plus;1&rp; &plus;1 &equ; &lp;a&plus;b&rp; &plus;
  &lp;1&plus;1&rp;</seg></seg></s> <emph rend="bl_1"/>
 <s type="es"><seg type="notation" ana="maths_arithmetic, algebra" rend="literal">
  <seg type="notation" ana="p" rend="literal">5&plus;&lp;6&plus;1&rp; &equ; &lp;5&plus;6&rp; &plus;1</seg></seg><lb rend="hl"/>
  <seg type="notation" ana="maths_arithmetic, algebra" rend="literal">
   <seg type="notation" ana="p" rend="literal">5&plus;&lp;6&plus;2&rp; &equ;</seg></seg></s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,12r[4]et12v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><c type="k">I</c> show you a curve drawn in a penta<lb rend="shyphen"/>gon which you had never
  thought of<lb/> &amp.and; I say&colon; I am sho<corr type="trsn"><orig type="trsn1">v</orig><reg type="trsn2">w</reg></corr>ing you that 
  <pb facs="Ms-148_12v" rend="verso" n="pagename_Ms-148,12v pageref_Ms-148,26"/><fw type="pageno" place="top left">11&p.ordinal;</fw> this curve <emph rend="us1">can</emph> be
  drawn,&dash; or&colon; that there<lb/> is such a curve in the pentagon&p.es;</s><lb rend="hl"/> 
 
 <s type="es" rend="indl_2">That there <emph rend="us1">are</emph> two twos in four&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,12v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Is there really no way out of saying,<lb/> say, that a triangle which has<lb/>
  <del type="dnpc">the</del> 3 equal sides has also three<lb/> equal angles&p.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,12v[3]et13r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Does <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-a.bmp" ana="graphics_Reihenfolgen; Gabelnotation" rend="bitmap">148026</seg> consist of
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-b.bmp" ana="graphics_Reihenfolgen; Gabelnotation" rend="bitmap">148027</seg> &amp.and;
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-c.bmp" ana="graphics_Reihenfolgen; Gabelnotation" rend="bitmap">148028</seg>&qm.eis;</s><lb/> 
 <s type="es">It depends what kind of dispute<lb/> it is&p.es;</s> 
 <s type="es">You could say it con<lb rend="shyphen"/>sisted of <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-d.bmp" ana="graphics_Reihenfolgen; Gabelnotation" rend="bitmap">148029</seg> &amp.and;
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-e.bmp" ana="graphics_Reihenfolgen; Gabelnotation" rend="bitmap">148030</seg>&p.es;</s> 
 <s type="es">Is the <del type="dnpc"><corr type="npcn"><gap extent="words_1"/></corr></del> dispute<lb/> one about facts or mode<lb/> of
  description&qm.eis;</s><lb rend="hl"/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-f.bmp" ana="graphics_Kreise; Kreis" rend="bitmap">148031</seg></emph>
 <s type="es">Counting  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-g.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">k148004</seg></emph>
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-h.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">k148005</seg></emph>
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-12v-i.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">k148006</seg></emph> 
  <pb facs="Ms-148_13r" rend="recto" n="pagename_Ms-148,13r pageref_Ms-148,27"/><fw type="pageno">12</fw>
  <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-0.bmp" ana="check_this bitmap should contain the four figures in  lines 1 through 5" rend="bitmap">m148000</seg> <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-a.bmp" ana="check_this bitmap should contain the four figures in  lines 1 through 5" rend="bitmap">m148003</seg></emph></s> 
  <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-b.bmp" ana="check_this bitmap should contain only the figure to  the right of the text and that looks like two aligned  rulers" rend="bitmap">m148004</seg></emph> <c type="c">T</c>his
   doesn<corr type="tra">&app.contr;</corr>t<lb/> show that<lb/> <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal"><seg type="notation" ana="p" rend="literal">a&plus;b</seg></seg> fits but<lb rend="hl"/> it shows that it
   looks like it does<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-c.bmp" ana="check_the bitmap contains only the pentagon within a star" rend="bitmap">m148005</seg></emph>
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-d.bmp" ana="check_the bitmap contains the star in a pentagon and the figure that overlaps with it" rend="bitmap">m148006</seg>
  <s type="es">&udq.sudq;What kind of figure do you<lb/> get if you draw the diago<lb rend="shyphen"/>nals
   in a <c type="c">P</c>entagon&qm.eis;&udq.eudq;</s><lb/> 
  <s type="es"><add rend="im">What sort of body do you get if you <corr type="trs"><orig type="trs1">&sdash;</orig> <reg type="trs2">draw the
   diagonals on a dodecaeder</reg></corr>&p.eis;</add></s><lb rend="hl"/>
  <s type="es">What kind of <abbr corresp="number">N<emph rend="sup">r</emph></abbr> do you get<lb/> if you draw
   <seg type="notation" ana="p" rend="literal">3s</seg> in <unclear>9</unclear>&p.eis;</s><lb rend="hl"/> <emph rend="bl_1"/> 
  <s type="es" rend="indl_2">What kind of <corr type="trs"><orig type="trs1">coulour</orig> <reg type="trs2">colour</reg></corr> do you get<lb/> if you
   <choice type="o"><orig type="o1">wet</orig><orig type="o2">mix</orig></choice> red with yellow&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/> 
  <s type="es" rend="indl_7"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13r-e.bmp" ana="graphics_Technische Darstellung; geometrischer  Beweis" rend="bitmap">148032</seg></emph></s> 
  <s type="es">&udq.sudq;The figure shows him that a pentagram<lb/> fits into a
   pentagon&udq.eudq;&p.es;</s><lb/> 
  <s type="es">Is this an experimental result&qm.eis;</s> </s> <lb rend="hl"/>
 <pb facs="Ms-148_13v" rend="verso" n="pagename_Ms-148,13v pageref_Ms-148,28"/><fw type="pageno" place="top left">13&p.ordinal;</fw></ab>
   


<ab n="Ms-148,13v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 
 <s type="es" rend="indl_2"><c type="c">I</c> am now talking <add rend="im">always</add> of a particular<lb/> kind of
  demonstration; what one<lb/> might call a<del type="dn">n</del> visual demonstration&p.es;</s> <lb rend="hl"/>
 <emph rend="bl_2"/></ab>
 


<ab n="Ms-148,13v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">In what sense could I say that<lb/> I didn<corr type="tra">&app.contr;</corr>t know that the
  pentagram<lb/> fitted the pentagon&qm.eis;</s> 
 <s type="es">Could I<lb/> have imagined the opposite&qm.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Suppose I had imagined the<lb/> opposite in some sense then<lb/> in the same sense I
  could<lb/> still hold the opposite after<lb/> the demonstration&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,13v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">We are talking of the star of which<lb/> I did <emph rend="us1">not</emph> know whether it
  fitted<lb/> the pentagon or not<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="bl_2"/> 
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13v-a.bmp" ana="graphics_Vielecke; F&uuml;nfeck" rend="bitmap">148033</seg> <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-13v-b.bmp" ana="graphics_Vielecke; Stern mit f&uuml;nf Punkten" rend="bitmap">148034</seg></s> <lb rend="hl"/>
 <s type="es">Then I try &amp.and; find that they<lb/> fit&p.es;</s> 
 <s type="es">This is just like making<lb/> an experiment&p.es;</s> <lb rend="hl"/><emph rend="bl_2"/>
 <pb facs="Ms-148_14r" rend="recto" n="pagename_Ms-148,14r pageref_Ms-148,29"/><fw type="pageno">14</fw></ab>
     


<ab n="Ms-148,14r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_2"><del type="dnpc"/> &udq.sudq;I never knew that I could<lb/> see the <c type="c">P</c>entagon
  &amp.and; its diagonals<lb/> in <emph rend="us1">this</emph> aspect&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14r-a.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148035</seg></s> <lb rend="hl"/>
 <s type="es" rend="indl_4">&udq.sudq;Oh, that&app.contr;s how it fits&em.ees;&udq.eudq;</s><lb rend="hl"/> 
 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14r-b.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148036</seg></s> <lb rend="hl"/>
 <pb facs="Ms-148_14v" rend="verso" n="pagename_Ms-148,14v pageref_Ms-148,30"/><fw type="pageno" place="top left">15</fw></ab>
  


<ab n="Ms-148,14v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14v-a.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148037</seg></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;<c type="c">I</c> don<corr type="tra">&app.contr;</corr>t know whether the
  <del type="dnpc"><corr type="npcn">dra</corr></del> penta<lb rend="shyphen"/>gram fits the pentagon&p.es;</s> 
 <s type="es">If so the<lb/> diagonals of a pentagon must give<lb/> a pentagram&p.es;</s> 
 <s type="es">Let&app.contr;s try it&p.es;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_2"/> 
 <s type="es" rend="indl_4">Is to see<del type="dn"><gap extent="characters_1"/></del> the figure<lb/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14v-b.bmp" ana="graphics_Vielecke; F&uuml;nfeck" rend="bitmap">148038</seg></emph> an experiment&qm.eis;</s> 
 <s type="es">Or to<lb rend="hl"/> see side by side the figures<lb rend="hl"/>
  <emph rend="slilm"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14v-c.bmp" ana="graphics_Vielecke; F&uuml;nfeck" rend="bitmap">148039</seg></emph>&qm.eis; 
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-14v-d.bmp" ana="graphics_Vierecke; Quadrat mit Punkt oben" rend="bitmap">148040</seg><lb rend="hl"/>
  <s type="es">But doesn&app.contr;t it teach us something&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/> 
  <s type="es" rend="indl_2"><emph rend="us1">&udq.sudq;It never struck me&udq.eudq;</emph></s> </s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,14v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">It seems we are learning by experience<lb/> a timeless truth about the shape of
  a<lb/> <c type="c">P</c>entagon &amp.and; of a <c type="c">P</c>entagram&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,14v[3]et15r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="c">I</c> never knew that one could look at it
  <pb facs="Ms-148_15r" rend="recto" n="pagename_Ms-148,15r pageref_Ms-148,31"/><fw type="pageno">16</fw> that way&p.es;</s> 
 <s type="es"><c type="k">I</c> had never seen the pentagram<lb/> in the pentagon&p.es;&udq.eudq;</s><lb rend="hl"/> 
<emph rend="bl_1"/></ab>


<ab n="Ms-148,15r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_7">It is a new experience to me&p.es;</s> 
 <s type="es">But is<lb/> it the experience teaching me that the<lb/> pentagram fits the
  pentagon&qm.eis;</s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,15r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><del type="dnpc">This</del> <c type="c">W</c>e feel that <del type="dnpc">the</del> here are two
  <add rend="im">visual</add> indivi<lb rend="shyphen"/>dualities which <del type="dnpc">we see</del> in the third
  <choice type="o"><orig type="o1"><unclear>indiv</unclear></orig> <orig type="o2">picture</orig></choice><lb/> we see <del type="dnpc"><corr type="npcn">p</corr></del> combined
  &amp.and; that we see that<lb/> they are capable of this particular
  combi<lb rend="shyphen"/>nation&p.es; <lb rend="hl"/><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-15r.bmp" ana="graphics_Dreiecke; Eckpunkt unten" rend="bitmap">148041</seg><lb rend="hl"/>
  <s type="es">The fact is that the combination &lp;not<lb/> meaning <c type="c">R</c>elation but
   <c type="c">C</c>omplex&rp; strongly<lb/> strikes us&p.es;</s> </s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,15r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><emph rend="slilm">&udq.sudq;The visual image <choice type="o"><orig type="o1"><unclear><seg type="notation" ana="p" rend="literal">P</seg></unclear></orig> <orig type="o2"><seg type="notation" ana="p" rend="literal">p</seg></orig></choice> fits
  the visual image <seg type="notation" ana="p" rend="literal">P</seg>&p.es;&udq.eudq;</emph></s><lb/> 
 <s type="es">The importance of this <abbr corresp="proposition">prop</abbr> lies in this<lb/> that
  it seems a <abbr corresp="proposition">prop&p.abb;</abbr> of experience<lb/> &amp.and;
  that on the other hand it also is<lb/> used as a
  <abbr corresp="proposition">prop&p.abb;</abbr> of geometry
  <abbr type="abb">i&p.abb;e&p.abb;</abbr> of grammar&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,15r[5]et15v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">What is the use of the <abbr corresp="proposition">prop&p.abb;</abbr> that
  <seg type="notation" ana="p" rend="literal">p</seg> fits <seg type="notation" ana="p" rend="literal">P</seg>&qm.eis;</s> 
 <pb facs="Ms-148_15v" rend="verso" n="pagename_Ms-148,15v pageref_Ms-148,32"/><fw type="pageno" place="top left">17</fw>
 <s type="es">&udq.sudq;If you draw the diagonals in <seg type="notation" ana="p" rend="literal">P</seg> you<lb/> get
  <seg type="notation" ana="p" rend="literal">p</seg>&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_7"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-15v.bmp" ana="graphics_Abbild; Gesicht&lp;er&rp;" rend="bitmap">148042</seg>
  &udq.sudq;<c type="c">I</c>f you do this &amp.and; this &amp.and; this &amp.and;
  this<lb/> you get <persName key="Napoleon Bonaparte" corresp="commentary">Napoleon</persName>&p.es;&udq.eudq;</s> <lb rend="hl"/><lb rend="hl"/></ab>
 


<ab n="Ms-148,15v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Problem&colon; &udq.sudq;<c type="c">D</c>raw that <c type="c">S</c>tar which will<lb/> fit the
  <c type="c">P</c>entagon&p.es;&udq.eudq;</s> 
 <s type="es">This is a mathe<lb rend="shyphen"/>matical problem&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,15v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;What do the diagonals of a <seg type="notation" ana="p" rend="literal">P</seg> look<lb/>
  like&qm.eis;&udq.eudq;</s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,15v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">We look at a puzzle picture &amp.and; find a<lb/> man in the foliage of a
  tree&p.es;</s> 
 <s type="es">Our<lb/> visual impression changes&p.es;</s> 
 <s type="es">But<lb/> <choice type="s"><orig type="alt1">can&app.contr;t</orig>  <orig type="alt2"><add rend="i">mustn&app.contr;t</add></orig></choice> we say that the new
  experi<lb rend="shyphen"/>ence would have been <emph rend="us1">impossible</emph><lb/> if the old one
  hadn&app.contr;t been what it<lb/> was&qm.eis;</s> 
 <s type="es">Such that we seem bound<lb/> to say the new
  <abbr corresp="experience">exp&p.abb;</abbr> was already<lb/> preformed in the old
  one&p.es;</s> 
 <s type="es">Or that<lb/> I found something new which was<lb/> already in the essence of the<lb/>
  first picture&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,15v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">We seem to have demonstrated an<lb/> internal property of the old
 picture&p.es;</s> <lb rend="hl"/>
 <pb facs="Ms-148_16r" rend="recto" n="pagename_Ms-148,16r pageref_Ms-148,33"/><fw type="pageno">18</fw></ab> 
    


<ab n="Ms-148,16r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16r-a.bmp" ana="check_this bitmap could contain some of the text, thus allowing for permanent alignement of the arrows with the images" rend="bitmap">m148007</seg></s> 
 <add rend="iupm"><c type="c">D</c>emonstrating that this is contained</add><lb rend="hl"/> in this
 <s type="es">&udq.sudq;It is in the nature of this to contain this &p.es;&udq.eudq;</s><lb rend="hl"/> 
 </ab>
 


<ab n="Ms-148,16r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16r-b.bmp" ana="maths_arithmetic, algebra" rend="bitmap">m148008</seg> if you
  do this &amp.and; this <abbr type="abb">etc<corr type="tra">&p.abb;</corr></abbr> you get
  55<corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16r-c.bmp" ana="graphics_Technische Darstellung; geometrischer Beweis" rend="bitmap">148043</seg></s> </ab>
 

<ab n="Ms-148,16r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
 <s type="es" rend="indl_2">Die mathematische Frage&p.es;</s><lb rend="hl"/> 
 <s type="es"><seg xml:lang="english">Could the
  <persName key="Pythagoras" corresp="commentary"><abbr corresp="Pythagorean">Pythag&p.abb;</abbr></persName>
  <corr type="tra">theorem</corr> be <emph rend="us1">assumed</emph> instead<lb/> of being
  deduced&qm.eis;</seg></s><lb rend="hl"/><emph rend="bl_2"/> </ab>

<ab n="Ms-148,16r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16r-d.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148044</seg></s> <emph rend="bl_1"/>
 <pb facs="Ms-148_16v" rend="verso" n="pagename_Ms-148,16v pageref_Ms-148,34"/><fw type="pageno" place="top left">19</fw> </ab>


<ab n="Ms-148,16v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;It never struck me that this <lb/><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-a.bmp" ana="check_arrow and pentagon included in bitmao" rend="bitmap">m148009</seg></emph> was this
  <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-b.bmp" ana="check_arrow and star included in bitmao" rend="bitmap">m148010</seg></emph><lb rend="hl"/>
  <s type="es">Something timeless seems to have<lb/> struck us&colon; <c type="c">H</c>ow can the
   identity<lb/> of these entities strike us<corr type="tra">&qm.eis;</corr></s><lb rend="hl"/> 
  <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-c.bmp" ana="graphics_Technische Darstellung; geometrischer  Beweis" rend="bitmap">148045</seg></emph></s> <lb rend="hl"/>
  <s type="es">It never struck me that 5 consisted<lb/> of <seg type="notation" ana="maths_arithmetic,  algebra" rend="literal">3 &plus; 2</seg><corr type="tra">&p.es;</corr></s> </s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,16v[2]et17r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">To see five figures as <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-d.bmp" ana="maths_arithmetic representation" rend="bitmap">m148011</seg>
  <abbr corresp="figures">F&p.abb;</abbr> &plus; 2 <abbr corresp="figures">F&p.abb;</abbr>
  &p.es;</s> 
 <s type="es">I<corr type="trsn"><orig type="trsn1">t</orig><reg type="trsn2">f</reg></corr> 5<lb/> <emph rend="us1">is</emph> &equ;
  <seg type="notation" ana="maths_arithmetic" rend="literal"> 2 &plus; 3</seg> it
  can&app.contr;t mean anything to<lb/> see 5 a
  <seg type="notation" ana="maths_arithmetic" rend="literal">2 &plus;
  3</seg><corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-e.bmp" ana="graphics_Reihenfolgen; Strichnotation" rend="bitmap">148046</seg> <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-f.bmp" ana="graphics_Klammern; Klammernotation &lp;mehrfach unten&rp;" rend="bitmap">148047</seg></s> <lb rend="hl"/>
 <s type="es">You could <corr type="trs"><orig type="trs1">devid</orig> <reg type="trs2">divide</reg></corr> 5 into
  <seg type="notation" ana="maths_arithmetic" rend="literal">2 &plus; 3</seg> but not<lb/> into
  <seg type="notation" ana="maths_arithmetic" rend="literal">3 &plus; <unclear>3</unclear></seg> as
  you could 6<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="bl_1"/>
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-16v-g.bmp" ana="graphics_Reihenfolgen; Strichnotation" rend="bitmap">148048</seg>
  <c type="c">O</c>bstacle</s>  
    <pb facs="Ms-148_17r" rend="recto" n="pagename_Ms-148,17r pageref_Ms-148,35"/><fw type="pageno">20</fw>
 <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17r-a.bmp" ana="graphics_Reihenfolgen;&#xA;Strichnotation" rend="bitmap">148049</seg><lb rend="hl"/></ab>


<ab n="Ms-148,17r[2]et17v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The whole question is really&colon; &udq.sudq;can it<lb/> <emph rend="us1">strike</emph> you
  what a thing <emph rend="us1">is</emph>&qm.eis;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17r-b.bmp" ana="graphics_Technische Darstellung; geometrischer Beweis" rend="bitmap">148050</seg></s> <lb rend="hl"/>
 <s type="es">It seems you can find out some<lb rend="shyphen"/>thing about the <emph rend="us1">nature</emph> of a
  thing<lb/> by experience&p.es;</s> 
 <s type="es">About its internal nature&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4"><del type="dnpc">It is</del></s> <lb rend="hl"/>
 <s type="es" rend="indl_3">Thus <abbr type="abb">e&p.abb;g&p.abb;</abbr> a <emph rend="us1">similarity</emph> can strike<lb/> you; the
  fact that a complex<lb/> <emph rend="us1">contains</emph> a constituent; even 
  <pb facs="Ms-148_17v" rend="verso" n="pagename_Ms-148,17v pageref_Ms-148,36"/><fw type="pageno" place="top left">21</fw> identity of shape&p.es;</s> 
 <s type="es">two tunes fitting<corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,17v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;One can see immediately that<lb/> 4 consists of
  <seg type="notation" ana="maths_arithmetic" rend="literal">2 &plus;
  2</seg>&udq.eudq;&p.es;</s> 
 <s type="es">This is nonsense<lb/> if <seg type="notation" ana="maths_arithmetic" rend="literal">4 &equ; 2
  &plus; 2</seg><corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,17v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17v-a.bmp" ana="graphics_Zeichen; Handschrift" rend="bitmap">148051</seg></s><lb rend="hl"/>
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17v-b.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148052</seg> 
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17v-c.bmp" ana="graphics_Dreiecke;  Eckpunkt oben" rend="bitmap">148053</seg> 
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-17v-d.bmp" ana="graphics_Dreiecke;  Eckpunkt oben" rend="bitmap">148054</seg>
  </s> <lb rend="hl"/></ab>
  
 

<ab n="Ms-148,17v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2"><del type="dnpc"><seg type="notation" ana="logic_propositional  formula" rend="literal"><seg type="notation" ana="p" rend="literal">&tilde.neg;p</seg></seg></del></s> <emph rend="bl_1"/>
  <s type="es"><seg type="notation" ana="logic_propositional formula" rend="literal"><seg type="notation" ana="p" rend="literal">&tilde.neg;
   &lb;&tilde.neg;p &pmid.con; &tilde.neg;&lp;&tilde.neg;r &pmid.con;
   &tilde.neg;s&rp;&rb; &pmid.con; &tilde.neg;&lb;&tilde.neg;&lp;&tilde.neg;t
   &pmid.con; <unclear>&tilde.neg;</unclear>&tilde.neg;s &pmid.con;&tilde.neg;t&rp;&rp;
   &pmid.con; &tilde.neg;p&rb;</seg></seg></s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,17v[5]et18r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" wabmarks-secml="dobslash.sm"><seg type="wabmarks-secml">&dobslash.sm;</seg>
  <s type="es" rend="indl_2">What do I do when <emph rend="us1">I draw your atten<lb rend="shyphen"/>tion to a fact</emph> about,
   say, this for<lb rend="shyphen"/>mula&qm.eis;</s> 
  <s type="es">It seems I make you see<lb/> something about its essence&p.es;</s> 
  <s type="es">You get<lb/> a new experience; but this experience<lb/> , it seems, teaches you
   something about<lb/> the essence the internal nature of the<lb/> formula&p.es;</s> 
  <s type="es">It seems to teach you<lb/> a mathematical &lp;<add rend="el">or</add> logical&rp; truth<lb/>
   &amp.and; this does not seem to be a rule<lb/> of grammar but a truth about the 
   <pb facs="Ms-148_18r" rend="recto" n="pagename_Ms-148,18r pageref_Ms-148,37"/><fw type="pageno">22</fw> 
   nature of things&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,18r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">If I made an experiment with a certain<lb/> figure we can <add rend="im"><gap extent="words_1"/></add>
   imagine this or that result&p.es;</s><lb/> 
 <s type="es">But if I draw your attention to a<lb/> feature</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,18r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">It consists of &sp.sms; appears to have 1&rp;<lb/> a grammatical meaning 2&rp;
   a physical<lb/> meaning &amp.and; 3&rp; a meaning lying between<lb/> these
   two&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,18r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">We seem to learn something about<lb/> the very
   sense<corr type="tra">&div;</corr>datum&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,18r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Tribe <del type="dnpc"><corr type="npcn">describ</corr></del> describing
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-a.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148055</seg> as
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-b.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148056</seg><lb/> or
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-c.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148057</seg><corr type="tra">&p.es;</corr></s> <lb rend="hl"/>
 <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,18r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
  <s type="es">A certain symbolism will easily go with<lb/> a certain aspect of looking at a<lb/>
   thing&p.es;</s> </emph><lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,18r[7]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;They regard the square as a double<lb/> right
   angle&p.es;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,18r[8]et18v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es">&udq.sudq;<seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-d.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148058</seg> consists of
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-e.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148059</seg>
  &amp.and; <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-f.bmp" ana="graphics_Abbild;  verschiedene" rend="bitmap">148060</seg>&udq.eudq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;<seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-g.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148061</seg> consists of
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-h.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148062</seg>
  &amp.and; <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18r-i.bmp" ana="graphics_Abbild;  verschiedene" rend="bitmap">148063</seg>&udq.eudq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
  
 <s type="es" rend="indl_2">In one case you say that it 
  <pb facs="Ms-148_18v" rend="verso" n="pagename_Ms-148,18v pageref_Ms-148,38"/><fw type="pageno" place="top left">23</fw> consists if it is
   d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided&p.es;</s> 
  <s type="es">In the<lb/> other case you <add rend="im">seem to</add> say that the<lb/>
   und<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided <unclear>object</unclear> consists
   &lp;time<lb rend="shyphen"/>lessly&rp; if you have seen a<lb/> similar object
   d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">Or you say that the object is<lb/> d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided if you have
   d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided its<lb/> picture&p.es;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,18v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2"><c type="k">I</c> dispose about 5 soldiers I imagine<lb/> them &amp.and; say&colon;
   I&app.contr;ll sen<corr type="trsn"><orig type="trsn1">t</orig><reg type="trsn2">d</reg></corr> <seg type="notation" ana="graphics_Reihenfolgen;  Strichnotation" rend="literal">&bar;&bar;&bar;</seg> to this<lb/> place
   <corr type="tra">and</corr> <seg type="notation" ana="graphics_Reihenfolgen;  Strichnotation" rend="literal">&bar;&bar;</seg> to that&p.es;</s> 
  <s type="es">Have I<lb/> thereby d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vided them into
   <seg type="notation" ana="maths_arithmetic" rend="literal">2 &plus; 3</seg><lb/>
   <c type="c">S</c>oldiers &amp.and; seen that it&app.contr;s possible&p.eis;</s> <lb/>
 <s type="es">What if I had imagined<lb/> this picture <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18v-a.bmp" ana="graphics_Klammern;  Klammernotation &lp;mehrfach unten&rp;" rend="bitmap">148064</seg>
   &amp.and; said<lb/> I&app.contr;ll send <seg type="notation" ana="graphics_Reihenfolgen;  Strichnotation" rend="literal">&bar;&bar;&bar;</seg> to this place<lb/>
   &amp.and; <seg type="notation" ana="graphics_Reihenfolgen;  Strichnotation" rend="literal">&bar;&bar;&bar;</seg> to
   that&qm.eis;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,18v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">What happens if our attention is drawn<lb/> to something <lb rend="hl"/>
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18v-b.bmp" ana="??" rend="bitmap">113ML12</seg>
   <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-18v-c.bmp" ana="??" rend="bitmap">K148007</seg>
   
   </s> <lb rend="hl"/></ab>



<ab n="Ms-148,18v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">One couldn&app.contr;t call 0&phigh.dec;<emph rend="pabove">3</emph> a shorthand for<lb/>
  0&phigh.dec;333&sp.ses;&p.es;</s> 
 <s type="es">Except insofar as 0&phigh.dec;33&sp.sms; is also a shorthand<lb/> for
  0&phigh.dec;333&sp.ses;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,18v[5]et19r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Don&app.contr;t try to find a 4 in the develop<corr type="npcn">e</corr>ment<lb/>
  it&app.contr;s hopeless&em.ees;&udq.eudq;&dash;</s> 
 <s type="es">&udq.sudq;Don&app.contr;t multiply
  <seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="p" rend="literal">25 &x.xmult; 25</seg></seg>
  <pb facs="Ms-148_19r" rend="recto" n="pagename_Ms-148,19r pageref_Ms-148,39"/><fw type="pageno">24</fw> again &amp.and; again in the hope to find
  600; it&app.contr;s<lb/> hopeless&em.ees;&udq.eudq;</s> <lb rend="hl"/>
 <s type="es" rend="indl_3">What&app.contr;s it like to <emph rend="us1">try to find</emph> a<lb/> 4 in the
  develop<corr type="npcn">e</corr>ment of <seg type="notation" ana="maths_arithmetic, real analysis, series expansion" rend="literal">1 &colon.div; 3</seg>&qm.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">And what is it like to find a 4&p.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">What is the <add rend="im">importance <corr type="npc">&amp.and;</corr></add> <del type="dnpc">meaning</del> of
  the question&colon;<lb/> &udq.sudq;<c type="c">W</c>hat is it like&qm.eis;&udq.eudq; or
  &udq.sudq;<c type="c">W</c>hat is the verification&qm.eis;&udq.eudq;</s> </emph><lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,19r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Kein Kalk&uuml;l ist im &udq.sudq;Widerspruch mit<lb/> der Logik&udq.eudq;
  <abbr type="abb">d&p.abb;h&p.abb;</abbr> mit gewissen Regeln die<lb/> &uuml;ber allen andern
  stehen&p.es;</s> 
 <s type="es">Die Annahme<lb/> einer obersten Logik ist es, die <del type="dnpc">die
  <unclear>Philosophen</unclear></del><lb/> hier irref&uuml;hrt&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>


<ab n="Ms-148,19r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">What we should call finding a 4 in <seg type="notation" ana="maths_arithmetic, real analysis, series expansion" rend="literal">1&slash;3</seg><lb/>
  obviously depends upon the operations in<lb/> this case&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">What does it mean to imagine<lb/> getting a result from a
  cal<choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>ulation&qm.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">How far is this imagination to go&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;There isn&app.contr;t a 4 in the first million<lb/>
  places&udq.eudq;&dash; &udq.sudq;<c type="c">Y</c>ou&app.contr;ve got a quick way of<lb/>
  calculating that&em.ees;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19r[7]et19v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Imagine this operation&colon; <c type="c">A</c> decimal 
  <pb facs="Ms-148_19v" rend="verso" n="pagename_Ms-148,19v pageref_Ms-148,40"/><fw type="pageno" place="top left">25</fw> fraction constructed by multiplying<lb/> again
  &amp.and; again <seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="p" rend="literal">25 &x.xmult;
  25</seg></seg><lb rend="hl"/> &colon; 0&phigh.dec;625625625&sp.ses;</s> <lb rend="hl"/>
 <s type="es">Look for an 8 in it&em.ees;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4">&udq.sudq;You know that you will never<lb/> find an 8&udq.eudq;
  means&colon;<lb rend="hl"/><emph rend="bl_1"/> <emph rend="indl_2"/> &udq.sudq;<c type="c">D</c>on&app.contr;t try to
  d<corr type="trsn"><orig type="trsn1">e</orig><reg type="trsn2">i</reg></corr>vide 2476 without<lb/> <unclear>remainder</unclear> by 3
  it&app.contr;s hopeless&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">In which case is it hopeless to find<lb/> a particular result by a
  calculation&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Calculating is the process of imagining<lb/> a calculation&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> can hope to find an 8 in the <c type="c">P</c>roduct <del type="dnpc"/><lb/>
  <seg type="notation" ana="maths_arithmetic" rend="literal">
  <seg type="notation" ana="p" rend="literal">284&x.xmult;379</seg></seg>&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">To say &udq.sudq;it&app.contr;s hopeless to find a certain<lb/> result
  <emph rend="us1">really</emph> means&colon; our calculation<lb/> <emph rend="us1">has already shown it to
   be wrong</emph>&p.es;<seg type="edinst">&ast;</seg></s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Or&colon; we have a calculation <del type="dnpc">of</del> which<lb/> we make have that
  <add rend="i">opposite</add> result&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,19v[7]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">What is the <choice type="dsl"><orig type="alt1"><del type="d">65<emph rend="sup">th</emph></del></orig>  <orig type="alt2"><add rend="i">56<emph rend="sup">th</emph></add></orig></choice>
  place of <seg type="notation" ana="maths_arithmetic, real analysis, series expansion" rend="literal">1 &colon.div; 7</seg>&qm.eis;</s><lb/> 
 <s type="es">You can now say <add rend="im">it seems</add> what the<lb/> <seg type="notation" ana="power">10<emph rend="power">10</emph></seg> place
  &usq.susq;<emph rend="us2">will</emph>&usq.eusq; be&p.es;</s> <lb rend="hl"/>
 <pb facs="Ms-148_20r" rend="recto" n="pagename_Ms-148,20r pageref_Ms-148,41"/><fw type="pageno">26</fw></ab>
   


<ab n="Ms-148,20r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">How can one calculation anticipate<lb/> the result of another&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english"><seg type="edinst">&ast;</seg>
 <s type="es" rend="indl_2">Or&colon; <corr type="npcn">&udq.sudq;</corr><c type="c">O</c>ur
  <choice type="em"><orig type="em1">cal<choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>ulus<add rend="i">ation</add></orig>  <orig type="em2"><choice type="s"><orig type="alt1">calculus</orig>  <orig type="alt2">calculation</orig></choice></orig></choice>
  has already <emph rend="us1">decided<lb/> against it</emph>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">What does it mean&colon; to prophecy what one<lb/> will <emph rend="us1">correctly</emph>
  find&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-20r.bmp" ana="graphics_Schema; ohne Linien" rend="bitmap">148065</seg>
  <seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="p" rend="literal">29 &x.xmult; 34 &equ; 34
  &x.xmult; 29</seg></seg><lb rend="hl"/><emph rend="bl_2"/>
  <emph rend="us1"><seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="p" rend="literal">3102 &x.xmult;
  2331</seg></seg></emph> <emph rend="us1"><seg type="notation" ana="maths_arithmetic" rend="literal">
   <seg type="notation" ana="p" rend="literal">2331 &x.xmult; 3102</seg></seg></emph></s><lb rend="hl"/> <emph rend="bl_3"/></ab>
 

<ab n="Ms-148,20r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Das Bild &udq.sudq;Alle&udq.eudq; angewandt auf die<lb/> Unendlichkeit&p.es;</s><lb rend="hl"/>
 <emph rend="bl_1"/> </ab>


<ab n="Ms-148,20r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">To show mathematically that a 4<lb/> can be found is to describe what it is<lb/>
  like to find a 4&p.es;</s> 
 <s type="es">And to find a 4 is here<lb/> a process in space and time&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20r[7]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_4">&udq.sudq;Find, as the result of a calculation &udq.eudq;&amp.and;<lb/>
  &udq.sudq;<c type="c">F</c>ind, otherwise&udq.eudq;<corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,20r[8]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_2">In <seg type="notation" ana="maths_arithmetic, analysis, series expansion" rend="literal">1 &colon.div; 7</seg> gibt es ein endliches
  Problem &amp.und; ein<lb/> unendliches&p.es;</s><lb rend="hl"/> <lb rend="hl"/>
 <pb facs="Ms-148_20v" rend="verso" n="pagename_Ms-148,20v pageref_Ms-148,42"/><fw type="pageno" place="top left">27</fw></ab>
  


<ab n="Ms-148,20v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-20v-a.bmp" ana="check_put both the large and small graphic into one bitmap" rend="bitmap">m148012</seg>
  <seg type="notation" ana="maths_arithmetic representation" rend="literal"/></s><lb rend="hl"/> <emph rend="bl_1"/> <emph rend="indl_2"/> 
 <s type="es">Two processes of calculation<lb/> lead to the same result&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;What if they at some stage<lb/> did not lead to the same<lb rend="hl"/>
  result&udq.eudq;&dash;&p.es;</s> 
 <s type="es">&udq.sudq;That is impossible, we<lb/> couldn&app.contr;t imagine their not
  leading<lb/> to the <corr type="trsn"><orig type="trsn1">t</orig><reg type="trsn2">s</reg></corr>ame result&p.es;&udq.eudq;</s> 
 <s type="es">But<lb/> then the proof of their leading<lb/> to the same result showed us<lb/> what it
  was like to lead to the<lb/> same result&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,20v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-20v-b.bmp" ana="check_put all three graphics into one bitmap" rend="bitmap">m148013</seg></s> <lb rend="hl"/>
 <pb facs="Ms-148_21r" rend="recto" n="pagename_Ms-148,21r pageref_Ms-148,43"/><fw type="pageno">28</fw></ab>
 
 

<ab n="Ms-148,21r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-21r.bmp" ana="check_I recommend one bitmap for all this  stuff" rend="bitmap">m148014</seg></s> <lb rend="hl"/></ab>
  

<ab n="Ms-148,21r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
   <s type="es" rend="indl_2"><emph rend="slilm">The difficulty consists in this that it<lb/> here seems impossible to
    imagine any<lb rend="shyphen"/>thing but what really is the case&colon; <c type="c">A</c>nd</emph><lb/>
    that of course means nothing&em.ees;</s><lb rend="hl"/> 
   <s type="es" rend="indl_4">We don&app.contr;t seem to be able to<lb/> imagine finding a 4,
    <emph rend="us1">because</emph> there<lb/> is a three there&p.es;</s> 
   <s type="es">But then how are<lb/> we capable of <emph rend="us2">imagining</emph> to find<lb/> a 3 <emph rend="us1">as
    there is</emph> a 3 there&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
   
  

<ab n="Ms-148,21r[3]et21v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
 <s type="es" rend="indl_2">If I say I can&app.contr;t imagine a 4 
  <pb facs="Ms-148_21v" rend="verso" n="pagename_Ms-148,21v pageref_Ms-148,44"/><fw type="pageno" place="top left">29</fw> to result it means that the<lb/> calculation
    shows me what it<lb/> means to imagine a 3 to result<lb/> &amp.and; gives no sense
    to the <abbr corresp="proposition">prop&p.abb;</abbr> &udq.sudq;I imagine<lb/> a 4 to
  result&udq.eudq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/><emph rend="bl_1"/> </ab>
   
  

<ab n="Ms-148,21v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
   <s type="es" rend="indl_4"><seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">x&pow2; &plus;
    ax &plus; b &equ;0</seg></seg></s> 
   <s type="es">&udq.sudq;Solve this equation <add rend="im">algebraically</add>&em.ees;&udq.eudq;</s><lb rend="hl"/>
 <emph rend="bl_1"/></ab>
   
  

<ab n="Ms-148,21v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
   <s type="es" rend="indl_2">&udq.sudq;Do something that has an analogy<lb/> to
    &sp.ses;&p.es;&udq.eudq;</s><lb rend="hl"/> 
   <s type="es" rend="indl_3">But we can&app.contr;t be sure that<lb/> we shall not in the end give up<lb/> the
    idea of something being analogous<lb/> to &sp.ses;&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/> </ab>
  
  

<ab n="Ms-148,21v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
   <s type="es" rend="indl_2">The existen<corr type="trsn"><orig type="trsn1">s</orig><reg type="trsn2">ce</reg></corr> of <choice type="s"><orig type="alt1">a</orig>  <orig type="alt2"><add rend="i">something we call
    the</add></orig></choice> &usq.susq;solution&usq.eusq; seems<lb/> to show clearly, that there
    was<lb/> a clear <emph rend="uw1">&amp.and; definite</emph> problem&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
   
  

<ab n="Ms-148,21v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1">
   <s type="es" rend="indl_2">Sup<corr type="tran">p</corr>ose we said that a solution is<lb/> a solution only
    <del type="dn">in</del><corr type="trs"><orig type="trs1">sofar</orig> <reg type="trs2">so far</reg></corr> as it<lb/> could have been described
    before it was found&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
   
  

<ab n="Ms-148,21v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
   <s type="es" rend="indl_5">&udq.sudq;Solve <seg type="notation" ana="maths_arithmetic, algebra" rend="literal">
    <seg type="notation" ana="p" rend="literal">x&pow2; &plus; 2ab &plus; b&pow2; &equ; 0</seg></seg>&udq.eudq;</s><lb rend="hl"/> 
   <s type="es">&udq.sudq;Solve <seg type="notation" ana="maths_arithmetic, algebra" rend="literal">
    <seg type="notation" ana="p" rend="literal">x&pow2; &plus; 4x &plus; 5 &equ; 0</seg></seg>&udq.eudq;</s><lb rend="hl"/><emph rend="bl_3"/>  
 <pb facs="Ms-148_22r" rend="recto" n="pagename_Ms-148,22r pageref_Ms-148,45"/><fw type="pageno">30</fw></ab>
 
 
  

<ab n="Ms-148,22r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
 
   <s type="es" rend="indl_2">&udq.sudq;We can&app.contr;t imagine that <seg type="notation" ana="maths_arithmetic,   real analysis, series expansion" rend="literal">1&colon.div;7</seg>
    should<lb/> not repeat itself after the dividend<lb/> has <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>ome
    back&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-22r-a.bmp" ana="maths_arithmetic" rend="bitmap">k148007</seg>
  <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-22r-b.bmp" ana="maths_arithmetic" rend="bitmap">m148015</seg></s><lb rend="hl"/> </ab>
 
  

<ab n="Ms-148,22r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
   <s type="es" rend="indl_2">We have <unclear>two ways</unclear> of calculating<lb/> the
    <seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="power">10<emph rend="power">10</emph></seg><emph rend="sup">th</emph></seg>
    place &amp.and; we can&app.contr;t imagine<lb/> that they lead to different
    results&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
   

<ab n="Ms-148,22r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Ist es eine <choice type="o"><orig type="o1">b</orig><orig type="o2">B</orig></choice>est&auml;tigung hierf&uuml;r wenn<lb/> die beiden
  <unclear>Bemerkungen</unclear> in einem<lb/> bestimmten Fall &uuml;bereinstimmen&qm.eis;</s><lb rend="hl"/> 
 <emph rend="bl_1"/></ab>


<ab n="Ms-148,22r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Is it different to say &udq.sudq;they lead<lb/> to the same
  <choice type="o"><orig type="o1">R</orig><orig type="o2">r</orig></choice>esult&udq.eudq; &amp.and; &udq.sudq;they<lb/> <emph rend="us1">must</emph> lead
  to the same result&udq.eudq;&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,22r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Does it mean anything to &udq.sudq;prophecy&udq.eudq;<lb/> the result of a
  calculation&qm.eis;</s><lb rend="hl"/> <emph rend="sepline"/></ab>
 


<ab n="Ms-148,22r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">We say we can&app.contr;t imagine that the two<lb/> processes should <add rend="i">not</add>
  lead to <del type="dnpc">the <corr type="npcn">sa</corr></del><lb/> the same result&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_5">What does it mean, we can&app.contr;t imagine it&qm.eis;</s><lb rend="hl"/>  <emph rend="bl_2"/>
 <pb facs="Ms-148_22v" rend="verso" n="pagename_Ms-148,22v pageref_Ms-148,46"/><fw type="pageno" place="top left">31</fw></ab>
    
    


<ab n="Ms-148,22v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-22v.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148066</seg></s> <lb rend="hl"/></ab>
 

<ab n="Ms-148,22v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_2">
  <s type="es" rend="indl_2">Must we recognise <c type="c">P</c>eriodicity as<lb/> a proof that there will be no<lb/> 6 in
   the development of <seg type="notation" ana="maths_arithmetic, real analysis, series  expansion" rend="literal">1&colon.div;7</seg>&qm.eis;</s><lb rend="hl"/> 
 <emph rend="bl_2"/></ab>
 

<ab n="Ms-148,22v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
  <s type="es" rend="indl_2">&udq.sudq;How does it happen that
   <seg type="notation" ana="maths_arithmetic" rend="literal"> <seg type="notation" ana="p" rend="literal">3&x.xmult;4</seg></seg> is
   <seg type="notation" ana="maths_arithmetic" rend="literal">
   <seg type="notation" ana="p" rend="literal">4&x.xmult;3</seg></seg>&qm.eis;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>

<ab n="Ms-148,22v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">&udq.sudq;An dieser Stelle mu&szlig; eine Primzahl<lb/> kommen&udq.eudq;&dash;</s> 
 <s type="es">&udq.sudq;An dieser Stelle <choice type="s"><orig type="alt1">kommt</orig>  <orig type="alt2"><add rend="i">steht</add></orig></choice> eine<lb/>
  Primzahl&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>

<ab n="Ms-148,22v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_4">&usq.susq;Gibt es einen Zufall in der Mathematik&qm.eis;&usq.eusq;</s><lb rend="hl"/> 
 <emph rend="bl_1"/></ab>


<ab n="Ms-148,22v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><emph rend="us2">How</emph> does the returning to the<lb/> dividend show me the periodicity
  of<lb/> the quotient&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,22v[7]et23r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">We seem in one kind of thought to<lb/> make jumps in the other to fill in<lb/> step
  by step&p.es;</s> 
 <s type="es">And the latter process<lb/> seems to justify the former&p.es;</s> 
 <pb facs="Ms-148_23r" rend="recto" n="pagename_Ms-148,23r pageref_Ms-148,47"/><fw type="pageno">32</fw>
 <s type="es">&udq.sudq;<emph rend="us1">You</emph> <emph rend="us2">see</emph> it just leads to the same<lb/>
  result&em.ees;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_1"/>
 <s type="es" rend="indl_6"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23r-a.bmp" ana="graphics_Kurven; Spirale" rend="bitmap">148067</seg></s><lb rend="hl"/> </ab>
 

<ab n="Ms-148,23r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Denke an den Fall wenn man mehrere<lb/> Z&uuml;ge in einem Spiel zusammenzieht
  &amp.und;<lb/> etwa im Schach gar nicht erst mit<lb/> der ersten Position
  anf&auml;ngt&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>

<ab n="Ms-148,23r[3]et23v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23r-b.bmp" ana="graphics_Klammern; Klammernotation &lp;mehrfach oben&rp;" rend="bitmap">148068</seg></s><lb rend="hl"/>
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23r-c.bmp" ana="check_put all dots and dashes into one bitmap" rend="bitmap">k148008</seg></s><lb rend="hl"/> 
 <s type="es" rend="indl_2">&udq.sudq;Die Form &usq.susq;1 2 3 4 5&usq.eusq; <add rend="our">p</add>a&szlig;t auf die
  Form<lb/> <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23r-d.bmp" ana="graphics_Schema; ohne Linien" rend="bitmap">148069</seg>&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Was f&uuml;r ein Faktum ist das, da&szlig; die Reihen<lb rend="shyphen"/>folge das Resultat nicht
  &auml;ndert&p.eis;</s><lb rend="hl"/> <seg xml:lang="english">
 <s type="es">The process we are going through just<lb/> <emph rend="us1">does</emph> <del type="dnpc">to</del> lead
  to the same result;&dash; but<lb/> so far as it &udq.sudq;leads to the same
  result<corr type="tra">&udq.eudq;</corr> we<lb/> could imagine it to lead to a different 
  <pb facs="Ms-148_23v" rend="verso" n="pagename_Ms-148,23v pageref_Ms-148,48"/><fw type="pageno" place="top left">33</fw> result&p.es;</s> 
 <s type="es">And so far as we couldn<corr type="tra">&app.contr;</corr>t<lb/> imagine it to lead to a
  different<lb/> result it doesn<corr type="tra">&app.contr;</corr>t lead to any<lb/> result but shows
  what it&app.contr;s<lb/> like to lead to the same<lb/> result&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_0"><abbr type="abb">I&p.abb;e&p.abb;</abbr>&colon; <c type="c">I</c>f we lost at the
  <emph rend="us2"><c type="c">F</c>orms</emph><lb/> <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23v-a.bmp" ana="graphics_Schema; ohne Linien" rend="bitmap">148070</seg> &amp.and; 12345as
  equivalent<lb/> there ceases to be a question of<lb/> whether the two processes lead<lb/>
  to the same or to different<lb/> results &amp.and; the apparent experi<lb rend="shyphen"/>ment
  <del type="dnpc"/> serves only to show<lb/> what sort of fact we take<lb/> as the standard
  of our expression&p.es;</s> </seg> <lb rend="hl"/><emph rend="bl_2"/></ab>

<ab n="Ms-148,23v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-23v-b.bmp" ana="check_one bitmap for everything here" rend="bitmap">
  k148009</seg></s><lb rend="hl"/>
 <emph rend="bl_2"/><pb facs="Ms-148_24r" rend="recto" n="pagename_Ms-148,24r pageref_Ms-148,49"/><fw type="pageno">34</fw></ab>
  


<ab n="Ms-148,24r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <emph rend="indl_2"/><emph rend="slilm">
 <s type="es">&udq.sudq;How can you impose two rules on<lb/> your
  arithmeti<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr> unless you know<lb/> that they must lead to the
  same<lb/> result&qm.eis;&udq.eudq;</s> </emph><lb rend="hl"/> 
 <s type="es" rend="indl_2">You wish to say&colon; &udq.sudq;<c type="c">T</c>hese rules by<lb/>
  <corr type="trs"><orig type="trs1">there</orig> <reg type="trs2">their</reg></corr> very nature, lead to the<lb/> same
  result&p.es;&udq.eudq;</s> 
 <s type="es">And you would therefore<lb/> have recognised something about<lb/> the very nature of
  them&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Now it is time that you can<lb/> make a man look into the
  <choice type="s"><orig type="alt1">case</orig>  <orig type="alt2">working</orig></choice><lb/> of these rules; that is, you can prove something
  about them&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;You go through this way of thinking<lb/> &amp.and; then you go
  through another way<lb/> of thinking which <add rend="i">independently</add> leads to the same<lb/>
  result&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">123456 123456<lb rend="hl"/> <emph rend="indl_6"/> <emph rend="us1">2</emph> 2 2</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24r[5]et24v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">After you have seen that <seg type="notation" ana="maths_arithmetic, real analysis, series expansion" rend="literal">1000 &colon.div; 3</seg> <emph rend="us1">must</emph><lb/>
  lead to 333 is it a confirmation to calculate<lb/> it &amp.and; see what it
  does&qm.eis;</s> 
 <s type="es">Hadn&app.contr;t you calcula<lb rend="shyphen"/>ted it by &udq.sudq;seeing that it was
  333&udq.eudq;&qm.eis;</s> 
 <s type="es">And<lb/> what does it mean that one calculation<lb/> confirms the result of the
 other&qm.eis;</s> <lb rend="hl"/> 
 <pb facs="Ms-148_24v" rend="verso" n="pagename_Ms-148,24v pageref_Ms-148,50"/><fw type="pageno" place="top left">35</fw>
 <s type="es" rend="indl_2">If you first see that the two <abbr corresp="calculations">calc&p.abb;</abbr><lb/>
  <emph rend="us2">must</emph> lead to the same result<lb/> is it a confirmation to find that<lb/>
  they do&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/>
 <s type="es" rend="indl_6"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-24v.bmp" ana="graphics_Technische Darstellung; geometrischer Beweis" rend="bitmap">148071</seg></s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;If this goes on this way &amp.and; that<lb/> goes on that way they
  <emph rend="us1">must</emph><lb/> meet there&em.ees;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_6">25 25 25 25 &sdash; 16 times<lb rend="hl"/> 16 16 16 16 25 times<lb rend="hl"/> they must meet
  at the end&p.es;</s><lb rend="hl"/> <emph rend="slilm">
 <s type="es">&udq.sudq;Are you surprised that they meet&qm.eis;</s><lb/> 
 <s type="es">Didn&app.contr;t you know that they had to<lb/> meet&qm.eis;&udq.eudq;</s> 
 </emph><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,24v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> wasn&app.contr;t surprised I always followed<lb/> the
  <seg type="notation" ana="p" rend="literal">25s</seg> while going on with the <seg type="notation" ana="p" rend="literal">16s</seg>&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <emph rend="bl_8"/><pb facs="Ms-148_25r" rend="recto" n="pagename_Ms-148,25r pageref_Ms-148,51"/><fw type="pageno">36</fw> </ab>
   


<ab n="Ms-148,25r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25r-a.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148072</seg></s><lb rend="hl"/> <emph rend="indl_2"/><emph rend="slilm">
 <s type="es">Must a series of dots give<lb/> the same number counted this &rightarr;<lb/> way
  &amp.and; that &leftarr;&qm.eis;</s> 
 <s type="es">&lp;There are two cases&p.es;&rp;</s> </emph><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,25r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Can <choice type="s"><orig type="alt1">we</orig>  <orig type="alt2"><add rend="i">I</add></orig></choice> try whether the result is the<lb/>
  same&qm.eis;&dash;</s> 
 <s type="es">It se<corr type="trsn"><orig type="trsn1">a</orig><reg type="trsn2">e</reg></corr>ms, yes&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25r-b.bmp" ana="graphics_Technische Darstellung; Korrelation" rend="bitmap">148073</seg></s><lb rend="hl"/> 
 <emph rend="bl_1"/>
 <s type="es">&udq.sudq;Can you imagine the calculation
  <seg type="notation" ana="maths_arithmetic" rend="literal"> <seg type="notation" ana="p" rend="literal">16&x.xmult;34</seg></seg><lb/> to
  lead to something else &sp.ses;&qm.eis;&udq.eudq;&dash;</s> 
 <s type="es">&udq.sudq;Can<lb/> you imagine these two calculations<lb/> leading to different
  results&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,25r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_7"><seg type="notation" ana="maths_arithmetic, real analysis, series expansion" rend="literal"> 1&colon.div;7&equ;&sdash;</seg>
  &udq.sudq;<c type="c">T</c>he division must give the same<lb/> result as it gave <del type="dnpc">the
  <corr type="npcn"><gap extent="characters_1"/></corr></del> <emph rend="us1">be<lb rend="shyphen"/>fore</emph>&p.es;&udq.eudq;</s> 
 <s type="es" rend="indl_6"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25r-c.bmp" ana="check_if the bitmap includes the arrows it would be best placed here" rend="bitmap">m148016</seg></s>
 <s type="es">Can we try whether<lb/> it does&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,25r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Can we imagine the same calculation<lb/> to lead <choice type="dsl"><orig type="alt1"><del type="d">twice</del></orig>  <orig type="alt2"><add rend="i">the
  second time</add></orig></choice> to a different result<del type="dnpc"/>&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,25r[5]et25v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The question is really whether there<lb/> can be a &udq.sudq;must&udq.eudq; in
  a <abbr corresp="proposition">prop&p.abb;</abbr> about the 
  <pb facs="Ms-148_25v" rend="verso" n="pagename_Ms-148,25v pageref_Ms-148,52"/><fw type="pageno" place="top left">37</fw>
  nature of things&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,25v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;In the sense in which they <emph rend="us1">&usq.susq;must&usq.eusq;</emph><lb/>
  lead, <del type="dnpc">they don&app.contr;t <emph rend="us1">lead</emph></del> we can&app.contr;t say<lb/>
  they <emph rend="us1">do</emph> lead&p.es;</s><lb rend="hl"/> <emph rend="bl_2"/></ab>
 

<ab n="Ms-148,25v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25v-a.bmp" ana="check_image difficult to read" rend="bitmap">m148017</seg><lb rend="hl"/> <emph rend="indl_2"/>
  &udq.sudq;<c type="c">W</c>enn die &Uuml;berlegung richtig ist&udq.eudq;, so<lb/> mu&szlig; diese
  Rechnung zu demsel<lb rend="shyphen"/>ben Resultat f&uuml;hren&p.es;</s><lb rend="hl"/> <lb rend="hl"/></ab>
   


<ab n="Ms-148,25v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
 <s type="es" rend="indl_7"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25v-c.bmp" ana="check_include everything, including the maths, in the bitmap" rend="bitmap">k148010</seg></emph> <c type="c">S</c>ie f&uuml;hren
  unabh&auml;ngig<lb/> zu selben Resul<lb rend="shyphen"/>tat&p.es;</s><lb rend="hl"/> <lb rend="hl"/></ab>


<ab n="Ms-148,25v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Let&app.contr;s imagine that we possessed<lb/> <emph rend="us2">only</emph> the
  <emph rend="us1">second</emph> criteri<corr type="trsn"><orig type="trsn1">on</orig><reg type="trsn2">um</reg></corr> for<lb/> determining
  divisibility&em.ees;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25v-d.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148074</seg> <c type="c">B</c>ut there seems to<lb/>
  be the difference be<lb rend="shyphen"/>tween &udq.sudq;they lead&udq.eudq; &amp.and;<lb/>
  &udq.sudq;they must lead&udq.eudq;&p.es;</s><lb rend="hl"/> 
 <s type="es"><seg type="notation" ana="logic_alphanumeric string" rend="literal"><seg type="notation" ana="p" rend="literal"><del type="d">d</del> a <del type="d">b</del>
  b b c &equ; <del type="d">d</del> c b a b</seg></seg><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-25v-e.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148075</seg></s><lb rend="hl"/> 
 <pb facs="Ms-148_26r" rend="recto" n="pagename_Ms-148,26r pageref_Ms-148,53"/><fw type="pageno">38</fw></ab>
    
   

<ab n="Ms-148,26r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 
 <s type="es" rend="indl_2">Wir nehmen ein falsches Verh&auml;ltnis von<lb rend="hl"/> Proze<corr type="trsn"><orig type="trsn1">ss</orig><reg type="trsn2">&szlig;</reg></corr>
  &amp.und; Resultat an&p.es;</s><lb/> 
 <s type="es" rend="indl_4">Denn es hei&szlig;t nicht da&szlig; ein <add rend="im">gewisser</add>
  Pro<lb rend="shyphen"/>ze<corr type="trsn"><orig type="trsn1">ss</orig><reg type="trsn2">&szlig;</reg></corr> zu einem <add rend="im">bestimmten</add> Resultat
  f&uuml;hren <emph rend="us2">mu&szlig;</emph>&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>

<ab n="Ms-148,26r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german">
 <s type="es" rend="indl_2">Denn ein Proze<corr type="trsn"><orig type="trsn1">ss</orig><reg type="trsn2">&szlig;</reg></corr> <emph rend="us2">mu&szlig;</emph> nur dazu f&uuml;hren<lb/> da&szlig; er
  geschehen ist&p.es;</s><lb rend="hl"/> <emph rend="bl_2"/></ab>

<ab n="Ms-148,26r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german"><seg xml:lang="english">
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-26r.bmp" ana="graphics_F&auml;den; mehrfachgeteilt" rend="bitmap">148076</seg></emph></s><lb rend="hl"/> 
 <s type="es" rend="indl_2">We sometimes substitute for the des<lb rend="shyphen"/>cription of a result<lb/> the
  description of a result&p.es;</s> </seg><lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,26r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">Ich kann mir eine Blume auf gewisse<lb/> Weise gewachs<del type="dn"><gap extent="characters_1"/></del>en
  denken&p.es;</s> 
 <s type="es">Und das Wachstum<lb/> ist dann ein Proze<corr type="trsn"><orig type="trsn1">ss</orig><reg type="trsn2">&szlig;</reg></corr> dessen Ende der<lb/>
  Zustand der Blume ist&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>

<ab n="Ms-148,26r[5]et26v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
 <s type="es" rend="indl_2">In welchem Sinne ist es m&ouml;glich<lb/> nicht zu wissen wohin ein
  <emph rend="us1">mathematischer</emph><lb/> Vorgang f&uuml;hrt&p.eis;</s> 
 <s type="es">Man k&ouml;nnte antworten<lb/> es ist m&ouml;glich nicht zu wissen, wohin er<lb/> f&uuml;hren
  <emph rend="us1">wird</emph> aber nicht, nicht zu wissen<lb/> wohin er f&uuml;hrt&p.es;</s> <lb rend="hl"/>
 <emph rend="indl_6"/><seg xml:lang="english">
 <s type="es">In one sense you can&app.contr;t know the<lb/> process without knowing the
  result, as<lb/> th<choice type="o"><orig type="o1">is</orig><orig type="o2">e</orig></choice> result is the <emph rend="us1">end</emph> of the
  process&p.es;</s> 
  <s type="es">In 
   <pb facs="Ms-148_26v" rend="verso" n="pagename_Ms-148,26v pageref_Ms-148,54"/><fw type="pageno" place="top left">39</fw> the other you may know a process
   &amp.and; not<lb/> know the result&p.es;</s><lb rend="hl"/> </seg><emph rend="bl_1"/></ab>


<ab n="Ms-148,26v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">In mathematics we <unclear>object</unclear> to say<lb/> these processes have the
  <del type="dnpc">end in<lb/> <corr type="npcn">com</corr></del></s><lb rend="hl"/><lb rend="hl"/> </ab>
 


<ab n="Ms-148,26v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-26v-a.bmp" ana="graphics_Vierecke; Quadrat" rend="bitmap">148077</seg></s><lb rend="hl"/> 
 <emph rend="bl_2"/></ab>


<ab n="Ms-148,26v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">A calculation leads to a result <add rend="im">mathematically</add><lb/> ap<corr type="npcn">p</corr>art
  from the fact whether I have<lb/> actually performed it&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,26v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&usq.susq;If I say this calculation <emph rend="us1">must</emph><lb/> lead to this result it
  has <del type="dnpc"><corr type="npcn">al</corr></del><lb/> already le<corr type="npcn">a</corr>d to
  it&p.es;&usq.eusq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,26v[6]et27r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-26v-b.bmp" ana="graphics_Reihenfolgen; Reihe" rend="bitmap">m148018</seg></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;<c type="k">I</c> knew it <emph rend="us1">beforehand</emph> what it must 
  <pb facs="Ms-148_27r" rend="recto" n="pagename_Ms-148,27r pageref_Ms-148,55"/><fw type="pageno">40</fw> lead to&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,27r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">If I say &usq.susq;this calculation must<lb/> lead to the same result&usq.eusq;
  by &udq.sudq;this<lb/> calculation<corr type="tra">&udq.eudq;</corr> I am referring to
  <add rend="im">whatever I call</add> a<lb/> method of calculating&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,27r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">Does calculating that there &dash;isn&app.contr;t<lb/> a six &sp.sms; confirm
  the result that<lb/> there couldn&app.contr;t be&qm.eis;</s> </emph><lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,27r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;You already see what happens, it<lb/> must always go on like
  this&p.es;&udq.eudq;</s><lb/> 
 <s type="es">Now suppose you actually went on<lb/> would this confirm what you saw<lb/>
  before&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,27r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">A man says, &udq.sudq;I see that the two calculations<lb/> <add rend="im">so far</add> agree
  but I don&app.contr;t know why they should<lb/> go on agreeing&udq.eudq;&p.es;</s>
 
 <s type="es"><del type="dnpc">In </del> <c type="c">S</c>hall we say that<lb/> he doesn&app.contr;t see a truth
  which the<lb/> other sees&qm.eis;&dash;</s> 
 <s type="es">He tries always again &amp.and; again&p.es;</s><lb/> 
 <s type="es">We ask him&colon; &udq.sudq;<c type="c">B</c>ut don&app.contr;t you see that<lb/> you
  must get to the same result again&qm.eis;<corr type="tra">&udq.eudq;</corr></s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Should we say that he must <del type="dnpc">make</del> go<lb/> the long way of experience,
  where we<lb/> go the shorter one of seeing&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/>
 <pb facs="Ms-148_27v" rend="verso" n="pagename_Ms-148,27v pageref_Ms-148,56"/><fw type="pageno" place="top left">41</fw></ab>
     
    


<ab n="Ms-148,27v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;If the multiplication le<corr type="npcn">a</corr>d to this<lb/> result once, it
  <emph rend="us1">must</emph> lead to <choice type="s"><orig type="alt1">it</orig>  <orig type="alt2"><add rend="i">the same result</add></orig></choice>
  again&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,27v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;What is the criteri<corr type="trsn"><orig type="trsn1">um</orig><reg type="trsn2">on</reg></corr> of
  periodicity&qm.eis;&udq.eudq;</s><lb/> 
 <s type="es">Here we are inclined to think that<lb/> we have a criterion the reappearance<lb/> of
  the remainder &amp.and; the actual periodi<lb rend="shyphen"/>city
  <add rend="i"><abbr type="abb">i&p.abb;e&p.abb;</abbr></add>, the repetition <seg type="ct">ad
  <abbr corresp="infinitum">inf<corr type="tra">&p.abb;</corr></abbr></seg> of the period&p.es;</s><lb rend="hl"/> 
<emph rend="bl_1"/></ab>


<ab n="Ms-148,27v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The infinite &amp.and; the huge&p.es;</s> 
 <s type="es">Absolute idea<lb/> of large &amp.and; small&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,27v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_4"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-27v.bmp" ana="check_include all here" rend="bitmap">k148011</seg></s><lb rend="hl"/> 
 <s type="es">&udq.sudq;<c type="k">I</c> never looked at it this way,
  before&udq.eudq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/><emph rend="bl_3"/>  
 <pb facs="Ms-148_28r" rend="recto" n="pagename_Ms-148,28r pageref_Ms-148,57"/><fw type="pageno">42</fw></ab>
    
   


<ab n="Ms-148,28r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">

 <emph rend="indl_2"/><emph rend="slilm">
 <s type="es">&udq.sudq;These people don&app.contr;t see a simple<lb/> truth
  &sp.ses;&p.es;<corr type="tra">&udq.eudq;</corr></s> </emph><lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,28r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">They are resolved to write<lb/> this&colon; <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-28r-a.bmp" ana="check_is bitmap of mathlog stuff here OK, bearing in mind the comparison with the next line" rend="bitmap"> k148012</seg> instead<lb rend="hl"/>
 of<lb/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-28r-b.bmp" ana="check_is bitmap of mathlog stuff here OK, bearing in mind the comparison with the line above" rend="bitmap"> k148013</seg></emph></s><lb rend="hl"/></ab>
 


<ab n="Ms-148,28r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">But not &udq.sudq;because it had to lead<lb/> <corr type="trs"><orig type="trs1">thru</orig> <reg type="trs2">through</reg></corr> to
  the same result&udq.eudq;&p.es;</s><lb rend="hl"/>  <emph rend="bl_2"/></ab>



<ab n="Ms-148,28r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
 <s type="es" rend="indl_2">It is a remarkable fact that people<lb/> almost always agree how to
  count&p.es;</s><lb rend="hl"/>  <emph rend="bl_2"/></ab>



<ab n="Ms-148,28r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Suppose I said this is the 100<emph rend="sup">th</emph><lb/> house of this street, although
  there<lb/> before are <del type="dnpc">no</del> only 5 houses built&p.es;</s><lb rend="hl"/> <emph rend="bl_3"/>
 <pb facs="Ms-148_28v" rend="verso" n="pagename_Ms-148,28v pageref_Ms-148,58"/><fw type="pageno" place="top left">43</fw></ab>
     


<ab n="Ms-148,28v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" wabmarks-secml="slash.sm">
 
 <seg type="wabmarks-secml">&slash.sm;</seg>
 <s type="es" rend="indl_2"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-28v-a.bmp" ana="check_include the strichnotation and the 1234,4321 number correlation" rend="bitmap">k148014</seg><emph rend="endarrback"/> <emph rend="endarrback"/>  <emph rend="endarrback"/> <c type="c">A</c>m
  I to be guided by this<emph rend="startarrback"/><lb/> or by this<emph rend="startarrback"/>&qm.eis;</s>  
 <s type="es">And how<lb/> do I know that they will<lb/> guide me to the same result&qm.eis;</s><lb rend="hl"/> 
 <s type="es">We have a general kind of ideas<lb/> of how it goes on; but can&app.contr;t<lb/>
  this after all be contradicted<lb/> by the actual detailed
  calcula<lb rend="shyphen"/>tion&qm.eis;</s> 
 <s type="es"><emph rend="us1">Isn&app.contr;t there a danger</emph> of it going<lb/> wrong after
  all&qm.eis;</s><lb rend="hl"/> 
 <s type="es" rend="indl_3">What is the truth which we<lb/> see &lp;&amp.and; which is
  &usq.susq;obvious&usq.eusq;&rp;&qm.eis;</s> 
 <s type="es">That</s> <lb rend="hl"/>
 <s type="es"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-28v-b.bmp" ana="check_include the strichnotation, the word This, quotation mark and the arrow Add the word This in hidden searchable text" rend="bitmap">k148015</seg><emph rend="endarrback"/></s><lb rend="hl"/> 
 <s type="es">This<emph rend="startarrback"/> shows us that this<emph rend="startarrback"/><lb/> was
  justified&udq.eudq;<corr type="tra">&p.es;</corr></s> 
 <s type="es" rend="indl_4"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-28v-c.bmp" ana="check_include the two vertical bars here" rend="bitmap">k148016</seg></emph> <c type="c">B</c>ut then<lb/> we
  leave behind us these<lb/> justifications&p.es;</s> 
 <s type="es">At first imagination<lb/> accompanies us a <del type="dnpc">&lp;certain&rp;</del> stretch<lb/>
  &amp.and; then <del type="dnpc">even</del> we are left alone&p.es;</s><lb rend="hl"/> </ab>
 


<ab n="Ms-148,28v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_4">If there are 777 in the first 100<lb/> places there are 777 in the infinite<lb/>
  develop<corr type="npcn">e</corr>ment&p.es;</s><lb rend="hl"/><emph rend="bl_1"/>  
 <pb facs="Ms-148_29r" rend="recto" n="pagename_Ms-148,29r pageref_Ms-148,59"/><fw type="pageno">44</fw></ab>
    
    


<ab n="Ms-148,29r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> have found 777 somewhere in<lb/> &pi;&udq.eudq;
  <del type="dnpc"><c type="c">T</c>his leaves</del><lb rend="hl"/> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-29r-a.bmp" ana="graphics_Kurven; Spirale" rend="bitmap">148078</seg></emph></s><lb rend="hl"/></ab>
 

<ab n="Ms-148,29r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
  <s type="es" rend="indl_2">&udq.sudq;The calculation guides you to<lb/> the result&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-29r-b.bmp" ana="graphics_Kurven;  Spirale" rend="bitmap">148079</seg></emph> &udq.sudq;<c type="c">Y</c>ou know
   that the<lb/> two rules <emph rend="us1">must</emph> <add rend="i">always</add> lead you<lb/> to the same
   result&p.es;&udq.eudq;</s><lb rend="hl"/> </ab>
 
 

<ab n="Ms-148,29r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0">
  <s type="es" rend="indl_2">The process of calculation <choice type="s"><orig type="alt1">can</orig>  <orig type="alt2"><add rend="i">may</add></orig></choice><lb/> be regarded as a
   process where<lb/> there is no compulsion or being guided<lb/> &amp.and; on the other
   hand, as a process<lb/> where <del type="dnpc">to</del> we move under some strict<lb/>
   guidance&p.es;</s><lb rend="hl"/> </ab>



<ab n="Ms-148,29r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-29r-c.bmp" ana="graphics_Technische Darstellung; geometrischer Beweis" rend="bitmap">148080</seg></emph></s><lb rend="hl"/> 
 <pb facs="Ms-148_29v" rend="verso" n="pagename_Ms-148,29v pageref_Ms-148,60"/><fw type="pageno" place="top left">45</fw></ab>
   


<ab n="Ms-148,29v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">&udq.sudq;If I follow this chain of<lb/> <emph rend="uw1">steps</emph> it&app.contr;s bound
  to lead me<lb/> there&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,29v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">&udq.sudq;&udq.sudq;The question are there 777 in <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg><lb/> is <corr type="trs"><orig type="trs1">allright</orig> <reg type="trs2">all right</reg></corr>
  because surely there either<lb/> are 777 in <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg> or
  there aren<corr type="tra">&app.contr;</corr>t&udq.eudq;&p.es;</s> 
 <s type="es">Queer&em.ees;<lb/> use of <seg type="notation" ana="logic_propositional formula" rend="literal"><seg type="notation" ana="p" rend="literal">p &vel; &tilde.neg;p</seg></seg>&p.es;</s> </emph>
 <s type="es">Images characteristic<lb/> for this statement&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_4">It real<corr type="tran">l</corr>y means&colon; &udq.sudq;<c type="c">T</c>he question is<lb/>
  <corr type="trs"><orig type="trs1">allright</orig> <reg type="trs2">all right</reg></corr> because there is a method<lb/> of verifying it
  although<del type="dn">t</del> we can&app.contr;t<lb/> use it&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,29v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;The third place of <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg> is 4<lb/>
  whether <emph rend="us1">I know it or</emph> not&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,29v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;What if we had proved it to be<lb/> self<corr type="tra">&div;</corr>contradictory
  that there should<lb/> be no 777 <add rend="im">in <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg></add>,
  mustn&app.contr;t we then say that<lb/> there are 777&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,29v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english"><emph rend="indl_2"/><emph rend="slilm">
 <s type="es">Our prose expressions in mathematics<lb/> are highly metaphorical&p.es;</s> 
</emph><lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,29v[6]et30r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Every algebraic <abbr corresp="equivalent">equ&p.abb;</abbr> has a
  root&udq.eudq;&p.es;</s><lb/> 
 <s type="es">Is this to be called a proposition&qm.eis;</s>  
 <pb facs="Ms-148_30r" rend="recto" n="pagename_Ms-148,30r pageref_Ms-148,61"/><fw type="pageno">46</fw> 
 <s type="es" rend="indl_2">The question corresponding to this<lb/> <abbr corresp="proposition">prop&p.abb;</abbr>
  as answer is vague&p.es;</s> 
 <s type="es">But once<lb/> the <abbr corresp="proposition">prop&p.abb;</abbr> this piece of
  mathematics has<lb/> been done we are inclined to call<lb/> it the proof that our
  question had<lb/> to be answered is the positive&p.es;</s> 
 <s type="es">But,<lb/> as one might say, there was much<lb/> less in question than there is<lb/> now
  in the answer&p.es;&dash;</s> 
 <s type="es">Compare<lb/> this with&colon; &udq.sudq;<choice type="o"><orig type="o1">i</orig><orig type="o2"><c type="c">I</c></orig></choice>s
  <seg type="notation" ana="maths_arithmetic" rend="literal"><seg type="notation" ana="p" rend="literal">25&x.xmult;25 &equ;
  600</seg></seg>&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2"><abbr corresp="Propositions"><c type="c">P</c>rops&p.abb;</abbr> which seem only to
  have sense<lb/> if their truth or falsehood is known&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">What kind of <abbr corresp="proposition">prop&p.abb;</abbr> will the
  <abbr corresp="proposition">prop&p.abb;</abbr><lb/> be that there can&app.contr;t &lp;or
  must&rp;<lb/> be 777 in <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Will it be possible <abbr type="abb">e&p.abb;g&p.abb;</abbr> to calculate<lb/> whether any
  given <abbr corresp="proposition">prop&p.abb;</abbr> of digits occurs<lb/> or how often
  it does&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">Relation between proof showing that<lb/> 777 must be between <seg type="notation" ana="p" rend="literal">n &amp.and;
  m</seg> &amp.and; <del type="dnpc">actual</del><lb/> <c type="c">P</c>roof that they are at the
  <seg type="notation" ana="p" rend="literal">v<emph rend="sup">th</emph></seg> place<lb/> &lp;<seg type="notation" ana="p" rend="literal">v</seg> bein<corr type="tran">g</corr> between <seg type="notation" ana="p" rend="literal">n
  &amp.and; m</seg>&rp;&p.es;</s><lb rend="hl"/> <emph rend="bl_3"/> 
 <pb facs="Ms-148_30v" rend="verso" n="pagename_Ms-148,30v pageref_Ms-148,62"/><fw type="pageno" place="top left">47</fw></ab>
    
    


<ab n="Ms-148,30v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
 <s type="es" rend="indl_2">Negation of a mathematical <abbr corresp="proposition">prop&p.abb;</abbr> of<lb/>
  fault in a calculation&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-30v.bmp" ana="graphics_Abbild; verschiedene" rend="bitmap">148081</seg></emph>
  <seg type="notation" ana="maths_arithmetic, algebra" rend="literal"><seg type="notation" ana="p" rend="literal">25&x.xmult;25 &equ;
  600x</seg></seg><lb rend="hl"/><emph rend="bl_1"/>
  <unclear><seg type="notation" ana="maths_trigonometry" rend="literal"><seg type="notation" ana="p" rend="literal"><gap extent="words_1"/>x
  &pmid.pmult; cos x &equ; sin x</seg></seg></unclear></s><lb rend="hl"/> <emph rend="bl_1"/> 
 <s type="es" rend="indl_2">&udq.sudq;Are there an infinite
  <abbr corresp="Number"><c type="c">N</c><emph rend="sup">r</emph></abbr> of 777 in<lb/>
  <seg type="notation" ana="maths_real analysis, transcendental number, series expansion" rend="literal">&pi;</seg>&qm.eis;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">&udq.sudq;There aren<corr type="tra">&app.contr;</corr>t&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Question&udq.eudq; corresponds to
  &udq.sudq;investiga<lb rend="shyphen"/>tion&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Heptagon must have been an<lb/> <c type="c">I</c>nvestigation&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">&udq.sudq;Is 5&p.ordinal; a cardinal
  <abbr corresp="number">N<emph rend="sup">r</emph></abbr>&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,30v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <s type="es" rend="indl_2">There is a contradiction between the<lb/> normal use of the word
  &udq.sudq;proposition&udq.eudq;<lb/> <corr type="tra">and</corr>
  &udq.sudq;question&p.es;<corr type="tra">&udq.eudq;</corr></s><lb rend="hl"/><emph rend="bl_4"/> 
 <pb facs="Ms-148_31r" rend="recto" n="pagename_Ms-148,31r pageref_Ms-148,63"/><fw type="pageno">11</fw></ab>
     
    


<ab n="Ms-148,31r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="vdline">
 
 <s type="es" rend="indl_2">&udq.sudq;Wouldn&app.contr;t one like to know with<lb/> <emph rend="us1">real</emph>
  certainty whether the other <choice type="dsf"><orig type="alt1">had</orig>  <orig type="alt2"><add rend="i"><del type="d">has</del></add></orig></choice><lb/>
  pains&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">Feeling of pastness&p.es;</s> 
 <s type="es">&udq.sudq;The experiences bound<lb/> up with the gesture <abbr type="abb">etc&p.abb;</abbr>
  aren&app.contr;t the<lb/> experience of pastness, for they could<lb/> be there without
  the feeling of past<lb rend="shyphen"/>ness&udq.eudq;&p.es;&dash;</s> 
 <s type="es"><corr type="npcn">&udq.sudq;</corr>But, on the other hand, would<lb/> it be that experience
  of pastness without<lb/> those experiences bound up with the<lb/>
  gesture&qm.eis;&dash;</s> 
 <s type="es"><choice type="o"><orig type="o1">w</orig><orig type="o2"><c type="c">W</c></orig></choice>hy should we say that<lb/> the
  <choice type="s"><orig type="alt1">characteristic</orig>  <orig type="alt2">&dobslash.alt;essential&dobslash.alt;</orig></choice> part is<lb/> the
  part outside those experiences&qm.eis;</s><lb/> 
 <s type="es">Isn&app.contr;t the experience at least partially<lb/> described if I have
  described the gestures<lb/> <abbr type="abb">etc<corr type="tra">&p.abb;</corr></abbr>&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2">Auch so&colon; <c type="c">D</c>ie Worte &udq.sudq;lang ist es her&dash;&udq.eudq;<lb/>
  rufen in mir manchmal ein bestimmtes<lb/> Gef&uuml;hl
  wach<choice type="o"><orig type="o1"><del type="dnpc">,</del></orig><orig type="o2">&p.es;</orig></choice></s> 
 <s type="es"><choice type="o"><orig type="o1">m</orig><orig type="o2"><c type="c">M</c></orig></choice>anchmal nicht&p.es;</s> 
 <s type="es">Aber wenn<lb/> sie es wachrufen so sind sie, ihr <gap extent="words_1"/><lb/> Teil der
  charakteristischen Erfahrung<corr type="tra">&p.es;</corr></s><lb rend="hl"/><emph rend="bl_2"/></ab>
 


<ab n="Ms-148,31r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2">Sprechen mit Andern &amp.und; mit mir selbst&colon;<lb/> &udq.sudq;<c type="c">W</c>enn
  ich eine gewisse Erfahrung habe, gebe<lb/> ich <add rend="im">&lp;nur&rp;</add> das Zeichen
  <add rend="i">&cross;</add>&sp.ses;&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <lb rend="hl"/> <pb facs="Ms-148_31v" rend="verso" n="pagename_Ms-148,31v pageref_Ms-148,64"/><fw type="pageno" place="top left">12&p.ordinal;</fw></ab>
  


<ab n="Ms-148,31v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">When one sa<corr type="trsn"><orig type="trsn1">i</orig><reg type="trsn2">y</reg></corr>s &udq.sudq;I talk to myself&udq.eudq;
  one<lb/> <add rend="im">generally</add> means just that one speaks &amp.and; is the<lb/> only
  person listening&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">If I look at something red <choice type="o"><orig type="o1">to</orig><orig type="o2">&amp.and;</orig></choice><lb/> say
  <del type="dnpc">&amp.and;</del>, to myself,<del type="dnpc"/> this is red, am I<lb/> giving
  myself an information&qm.eis;</s> 
 <s type="es">Am<lb/> I communicating a personal experience<lb/> to myself&p.es;</s> 
 <s type="es">Some <add rend="i">philosophising</add> people might be in<lb rend="shyphen"/>clined to say that this
  is the only real<lb/> case of communication of personal<lb/> experience because only I
  know<lb/> what I really mean by &usq.susq;red&usq.eusq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">Remember <del type="dnpc">that</del> in which special cases<lb/> only it has sense to
  inform <choice type="em"><orig type="em1">a<add rend="el">n</add> <add rend="i">other</add> per<lb rend="shyphen"/>son</orig>  <orig type="em2"><choice type="s"><orig type="alt1">a person</orig>  <orig type="alt2">an other
  person</orig></choice></orig></choice> that the colour he sees now<lb/> is red<del type="dn"><gap extent="characters_1"/></del>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">One doesn&app.contr;t say to oneself &udq.sudq;<choice type="o"><orig type="o1">t</orig><orig type="o2"><c type="c">T</c></orig></choice>his
  is a chair&p.es;&dash;</s> 
 <s type="es">Oh<lb/> really&qm.eis;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,31v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2">Wie kann ich denn einer Erfahrung
  &lp;<choice type="o"><orig type="o1"><abbr type="abb">z&p.abb;B&p.abb;</abbr></orig><orig type="o2">etwa</orig></choice><lb/> einem Schmerz&rp; einen
  Namen geben&qm.eis;</s> 
 <s type="es">Ist es<lb/> nicht als wollte ich ihm, etwa, einen Hut aufsetzen&qm.eis;</s> 
<emph rend="bl_1"/></ab>


<ab n="Ms-148,31v[6]et32r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2">Nehmen wir an man sagte&colon; &udq.sudq;<c type="c">M</c>an kann 
  <pb facs="Ms-148_32r" rend="recto" n="pagename_Ms-148,32r pageref_Ms-148,65"/><fw type="pageno">13</fw> ihm nur indirekt einen Hut
  aufsetzen&udq.eudq; so w&uuml;rde<lb/> ich fragen&colon; <c type="c">G</c>laubst Du da&szlig; man je
  auf die<lb/> Idee gekommen w&auml;re davon zu reden wenn<lb/> man nicht daran gedacht h&auml;tte
  da&szlig;<lb/> man dem Menschen der Schmerzen hat einen<lb/> Hut aufsetzen kann&qm.eis;</s> 
 <s type="es">Zu sagen man<lb/> k&ouml;nne dem Schmerz nur indirekt einen Hut<lb/> aufsetze macht es
  erscheinen als g&auml;be es<lb/> dennoch einen direkter Weg der nur tats&auml;ch<lb rend="shyphen"/>lich
  nicht gangbar sei&p.es;</s> 
 <s type="es"><gap extent="lines_1"/></s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,32r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">The difficulty is that we feel that we<lb/> have said something about the
  <add rend="our">na</add>ture<lb/> of pain when we say that one person<lb/> can&app.contr;t have
  another person<corr type="npc">&app.contr;</corr>s pain&p.es;</s><lb/> 
 <s type="es">Perhaps we shouldn&app.contr;t be inclined<lb/> to say that we had anything<lb/>
  physiological or even p<choice type="o"><orig type="o1">h</orig><orig type="o2">s</orig></choice>ychological<lb/> but something
  metapsychological<lb/> metaphysical&p.es;</s> 
 <s type="es">Something about<lb/> the essence, nature, of pain as<lb/> opposed to its causal
  connections to<lb/> other phenomena&p.es;</s><lb rend="hl"/><emph rend="bl_1"/></ab>


<ab n="Ms-148,32r[3]et32r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2">Es scheint uns etwa als w&auml;re es zwar<lb/> nicht falsch sondern unsinnig zu<lb/>
  sagen &udq.sudq;ich f&uuml;hle seine Schmerzen&udq.eudq;, aber<lb/> als w&auml;re dies so
  infolge der Natur 
  <pb facs="Ms-148_32v" rend="verso" n="pagename_Ms-148,32v pageref_Ms-148,66"/><fw type="pageno" place="top left">14</fw> des Schmerzes, der Person
  <abbr type="abb">etc&p.abb;</abbr>&p.es;</s> 
 <s type="es">Als<lb/> w&auml;re also jene Aussage letzten<lb/> Endes doch eine Aussage &uuml;ber<lb/> die Natur
  der Dinge&p.es;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2">Wir sprechen also etwa von<lb/> einer Asymmetrie<lb/> unserer Ausdrucks<lb rend="shyphen"/>weise
  &amp.und; fassen diese auf als ein<lb/> <choice type="o"><orig type="o1">s</orig><orig type="o2">S</orig></choice>piegelbild des Wesens der
  Dinge&p.es;</s><lb rend="hl"/> <emph rend="bl_2"/></ab>
 


<ab n="Ms-148,32v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">Intangibility of impressions&p.es;</s> 
 <s type="es">&lp;Anguish&rp;</s><lb/> 
 <s type="es">Some we should <gap extent="words_1"/> were more <unclear>tangible</unclear><lb/> than
  others&p.es;</s> 
 <s type="es">Seeing more tangible<lb/> than a faint pain; &amp.and; this more tangi<lb rend="shyphen"/>ble
  than a vague fear, longing <abbr type="abb">etc&p.abb;</abbr></s><lb rend="hl"/> 
 <s type="es" rend="indl_5">In what way are these intangi<lb rend="shyphen"/>ble experiences less easy to
  communicate<lb/> <del type="dnpc"><corr type="npcn">tha</corr></del> to describe than the
  <del type="dnpc"><corr type="npcn">mo</corr></del> &usq.susq;simpler&usq.eusq; ones&qm.eis;</s><lb rend="hl"/> 
 
 <s type="es" rend="indl_3">In what way do we use the &dash;phrase&colon;<lb/> &udq.sudq;<c type="c">T</c>his
  experience is difficult to describe&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <s type="es" rend="indl_2"><choice type="em"><orig type="em1">And can <add rend="im">it be even impossible to describe certain</add> <del type="d">an</del>
  experience<add rend="el">s</add></orig>  <orig type="em2"><choice type="dsl"><orig type="alt1"><del type="d"><c type="c">A</c>nd can an
  experience</del></orig>  <orig type="alt2"><c type="c">A</c>nd can it be even impossible to describe certain
  experiences</orig></choice></orig></choice><del type="d">be ever <unclear><corr type="npcn">in<lb rend="shyphen"/>des</corr></unclear></del>&qm.eis;</s><lb rend="hl"/>
 </ab>
 


<ab n="Ms-148,32v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german" emph="slilm">
 <s type="es" rend="indl_2"><del type="dnpc"><corr type="npcn">des</corr></del> <emph rend="indl_2"/> <c type="c">W</c>as f&uuml;r einen Sinn hat es zu sagen<lb/>
  diese <emph rend="uw1">Erfahrung</emph> ist nicht beschreib<lb rend="shyphen"/>bar&qm.eis;</s> 
 <s type="es">Wir m&ouml;chten sagen&colon; sie ist zu <corr type="trsn"><orig type="trsn1">c</orig><reg type="trsn2">k</reg></corr>om<lb rend="shyphen"/>plex, zu
  subtil&p.es;</s> <lb rend="shyphen"/><lb rend="hl"/>
 <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,32v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">&udq.sudq;Diese Erfahrung ist nicht mitteilbar, aber<lb/> <emph rend="us1">ich</emph> kenne
   sie,&dash; weil ich sie habe&p.es;&udq.eudq;</s><lb rend="hl"/>  
 <lb rend="hl"/>  <pb facs="Ms-148_33r" rend="recto" n="pagename_Ms-148,33r pageref_Ms-148,67"/><fw type="pageno">15</fw></ab>
   
 

<ab n="Ms-148,33r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">

  <s type="es" rend="indl_2">&udq.sudq;Es gibt die Erfahrung, &amp.und; die Beschreibung<lb/> der
   Erfahrung&p.es;&dash;<del type="dnpc">&udq.eudq;</del></s> 
  <s type="es">Daher kann es nicht<lb/> gleichg&uuml;ltig sein, ob der Andere die selbe<lb/> Erfahrung
   hat<del type="dnpc">,</del>, wie ich, oder nicht;&dash; &amp.und;<lb/> daher <choice type="dsl"><orig type="alt1"><del type="d">kann
   es</del></orig>  <orig type="alt2"> <add rend="i">mu&szlig; es</add></orig></choice> wenn ich mit mir selbst rede<lb/> auf
   <choice type="s"><orig type="alt1"><emph rend="uw1">diese</emph></orig>  <orig type="alt2"> <add rend="i">meine</add></orig></choice> Erfahrung ankommen&p.es;</s> 
  <s type="es"><del type="dnpc">&udq.sudq;</del>Es mu&szlig;<lb/> dabei eine entscheidende Rolle spielen<lb/> da&szlig;
   ich diese Erfahrung kenne &lp;w&auml;hrend ich<lb/> mit der des Andern nicht direkt
   vertraut bin&rp;&p.es;&udq.eudq;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,33r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Kann man sagen&colon; &udq.sudq;<choice type="o"><orig type="o1">i</orig><orig type="o2"><c type="c">I</c></orig></choice>n <choice type="s"><orig type="alt1">dem</orig>  <orig type="alt2">
   <add rend="i">das</add></orig></choice> was ich &uuml;ber<lb/> die Erfahrung des Andern sage, spielt seine<lb/>
   Erfahrung &lp;selbst&rp; nicht hinein&p.es;</s> 
  <s type="es">In <choice type="s"><orig type="alt1">dem</orig>  <orig type="alt2"> <add rend="i">das</add></orig></choice> was<lb/> ich &uuml;ber meine Erfahrung sage spielt
   <choice type="dsl"><orig type="alt1"><del type="d">sie</del></orig>  <orig type="alt2"> <add rend="i">diese Erfahrung</add></orig></choice><lb/> selbst
   hinein&p.es;&udq.eudq;&qm.eis;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">&udq.sudq;Ich spreche &uuml;ber meine Erfahrung,<lb/> sozusagen, <choice type="dsf"><orig type="alt1">in ihrer
   Anwesenheit&udq.eudq;</orig>  <orig type="alt2"> <del type="d">&dobslash.alt;in ihrem<lb/>
    Beisein&dobslash.alt;</del></orig></choice>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/>
</ab>
  

<ab n="Ms-148,33r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Wie wenn jemand sagen w&uuml;rde&colon; &udq.sudq;<c type="c">E</c>s gibt nicht<lb/> nur die
   Beschreibung des Tisches sondern<lb/> auch den Tisch&p.es;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_2"/></ab>
  
 

<ab n="Ms-148,33r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">&udq.sudq;Es gibt nicht nur das Wort &usq.susq;Zahnschmerz&usq.eusq;,<lb/>
   <choice type="s"><orig type="alt1">es gibt auch <choice type="s"><orig type="alt1"><add rend="i"><seg xml:lang="eng">such a thing as</seg></add></orig>  <orig type="alt2"> etwas wie</orig></choice> den
   Zahnschmerz<lb/> selbst&p.es;&udq.eudq;</orig>  <orig type="alt2"> &dobslash.alt;&sp.psa; es gibt auch
   Zahnschmerzen&p.es;&udq.eudq;&dobslash.alt;</orig></choice> </s><lb rend="hl"/>  <emph rend="bl_1"/> 
 <pb facs="Ms-148_33v" rend="verso" n="pagename_Ms-148,33v pageref_Ms-148,68"/><fw type="pageno" place="top left">16</fw></ab>
   
    
 

<ab n="Ms-148,33v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
 
  <s type="es" rend="indl_2">Es scheint, da&szlig;, da ich etwa<lb/> eine Erfahrung nicht beschreiben<lb/> kann, sie
   aber habe, da&szlig; ich<lb/> sie daher genauer <emph rend="us1">kennen</emph> kann,<lb/> als irgend ein
   Anderer&p.es;</s> 
  <s type="es">Aber was<lb/> hei&szlig;t, <del type="dnpc">es</del> die Erfahrung kennen,<lb/> wenn es nicht hei&szlig;t,
   sie beschreiben<lb/> &amp.und; nicht <add rend="i">hei&szlig;t</add>, sie haben&p.eis;</s> <lb rend="hl"/>
  
  <s type="es" rend="indl_4">Gibt es eine <emph rend="us2">Kenntnis</emph> der Erfah<lb rend="shyphen"/>rung, die wir nicht
   mitteilen k&ouml;nnen&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,33v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Hat es Sinn zu sagen &udq.sudq;ich kenne diese<lb/> Erfahrung <choice type="s"><orig type="alt1">besser</orig>  <orig type="alt2">
   &dobslash.alt;genauer&dobslash.alt;</orig></choice> als irgend ein<lb/> Anderer sie kennen
   kann&udq.eudq;&qm.eis;<del type="d">&udq.eudq;</del>&p.es;</s> 
  <s type="es">Gibt es Erfah<lb rend="shyphen"/>rungen die der Andere ebensogut kennen<lb/> kann wie ich
   &amp.und; solche, die er nicht so gut<lb/> kennen
   kann&qm.eis;<del type="dnpc">&dash;</del></s> 
  <s type="es">Hei&szlig;t das&colon; er kann diese<lb/> selbe <add rend="im">komplizierte</add> Erfahrung nicht
   haben&qm.eis;&dash;</s> 
  <s type="es">Es hei&szlig;t wohl&colon;<lb/> &udq.sudq;<c type="c">E</c>r kann sie haben, aber wir k&ouml;nnen
   <choice type="s"><orig type="alt1">nie</orig>  <orig type="alt2"> <add rend="i">nicht</add></orig></choice><lb/> wissen, da&szlig; er <choice type="s"><orig type="alt1">gerade</orig>  <orig type="alt2"> <add rend="i">genau</add></orig></choice>
   diese <del type="dnpc"/> gehabt hat&udq.eudq;&p.es;</s><lb/> 
  <s type="es"><abbr type="abb">Z&p.abb;B&p.abb;</abbr> <del type="dnpc"><gap extent="words_1"/></del> scheint es als k&ouml;nnten wir<lb/>
   sagen&colon; &udq.sudq;<c type="c">W</c>ir k&ouml;nnen in einem Sinn wissen da&szlig;<lb/> er gerade
   diese <choice type="o"><orig type="o1">E</orig><orig type="o2">e</orig></choice>inf&auml;rbige, glatte, <choice type="o"><orig type="o1">R</orig><orig type="o2">r</orig></choice>ote Fl&auml;che<lb/> sieht,
   aber nicht, da&szlig; er genau dieses <emph rend="us1">Flimmern</emph><lb/> sieht&p.es;</s> 
  <s type="es">Weil sich das genaue Gesichtsbild<lb/> <choice type="em"><orig type="em1">beim <add rend="i">des</add> Flimmern<add rend="i">s</add>
   </orig>  <orig type="em2"><choice type="s"><orig type="alt1">beim Flimmern</orig>  <orig type="alt2"> des Flimmerns</orig></choice></orig></choice> nicht beschreiben
   l&auml;&szlig;t&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,33v[3]et34r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Es gibt ja auch den Fall, in dem wir ein Gesichts<lb rend="shyphen"/>bild genauer durch
   ein gemaltes Bild als durch 
   <pb facs="Ms-148_34r" rend="recto" n="pagename_Ms-148,34r pageref_Ms-148,69"/><fw type="pageno">17</fw> Worte
   beschreiben k&ouml;nnen&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,34r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Wie ist es damit&colon; &udq.sudq;<c type="c">M</c>an kann eine <choice type="o"><orig type="o1">f</orig><orig type="o2">F</orig></choice>igur
   genauer<lb/> mit Hilfe von <choice type="o"><orig type="o1">m</orig><orig type="o2">M</orig></choice>a&szlig;zahlen als ohne diese
   be<lb rend="shyphen"/>schreiben&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,34r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Aber die Erfahrung, die ich <emph rend="us1">habe</emph> scheint <del type="dnpc"><add rend="i">im gewissen
   Sinne</add></del><lb/> eine Beschreibung dieser Erfahrung<del type="d_c">, im gewissen<lb/>
   Sinne,</del> zu ersetzen&p.es;</s> 
 <s type="es">&udq.sudq;Sie ist ihre eigene Beschreibung&udq.eudq;&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,34r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm"><emph rend="indl_2"/><emph rend="slilm">
  <s type="es">Vermischen wir hier nicht zwei Dinge&colon; die Zusam<lb rend="shyphen"/>mengesetztheit
   der Erfahrung &amp.und;, was man<lb/> ihren urspr&uuml;nglichen
   <choice type="s"><orig type="alt1"><emph rend="uw1">Geschmack</emph></orig>  <orig type="alt2"> <add rend="i"><emph rend="uw2">Ton</emph></add></orig> <orig type="alt3">
   <add rend="i"><seg xml:lang="eng">flavour</seg></add></orig></choice> nennen k&ouml;nnte&qm.eis;</s><lb/> 
 <s type="es">Ihre eigentliche nat&uuml;rliche Farbe<corr type="tra">&qm.eis;</corr></s> </emph><lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,34r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Es ist die Auffassung, da&szlig; von der urspr&uuml;ng<lb rend="shyphen"/>lichen Erfahrung
   <del type="dnpc">etwas</del> nur ein Teil<lb/> <choice type="s"><orig type="alt1">bei</orig>  <orig type="alt2"> <add rend="i">in</add></orig></choice> der Mitteilung
   erhalten bleibt, &amp.und; etwas<lb/> anderes <add rend="imw">von ihr</add> verloren
   geht&p.es;</s> 
  <s type="es">N&auml;mlich eben &usq.susq;ihr <seg xml:lang="english">timbre</seg>&usq.eusq;,<lb/> oder wie
   man es nennen m&ouml;chte&p.es;</s> 
  <s type="es">Es kommt<lb/> <corr type="trsn"><orig type="trsn1">E</orig><reg type="trsn2">e</reg></corr>inem hier so vor als k&ouml;nnte man,
   sozusa<lb rend="shyphen"/>gen nur die <add rend="im">farblose</add> Zeichnung vermitteln &amp.und; der
   Andere<lb/> setzte in sie <emph rend="us1">seine</emph> Farben ein&p.es;</s> 
 <s type="es">Aber das ist<lb/> nat&uuml;rlich &lp;<emph rend="uw1">eine</emph>&rp; T&auml;uschung&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,34r[6]et34v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Aber k&ouml;nnen wir nicht wirklich sagen, wir<lb/> h&auml;tten in dem Andern durch
   unsere Beschrei<lb rend="shyphen"/>bung ein Bild hervorgebracht aber wir<lb/> k&ouml;nnen nicht
   wissen ob dieses Bild nun 
   <pb facs="Ms-148_34v" rend="verso" n="pagename_Ms-148,34v pageref_Ms-148,70"/><fw type="pageno" place="top left">18</fw> genau das
   gleiche ist, wie das unsere&qm.eis;</s> <lb/>
  <s type="es">Denken wir hier an den Gebrauch<lb/> des Wortes
   <corr type="tra">&udq.sudq;</corr>gleich<corr type="tra">&udq.eudq;</corr> in <add rend="im">solchen</add> S&auml;tze
   wie&colon; &udq.sudq;<c type="c">D</c>iese<lb/> Kreise sind dem <emph rend="uw1">Augenschein</emph> nach<lb/>
   ganz gleich&p.es;&udq.eudq;</s><lb rend="hl"/> <emph rend="bl_1"/> </ab>
 
 

<ab n="Ms-148,34v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german" emph="slilm">
  <s type="es" rend="indl_2">Hierher geh&ouml;rt auch, da&szlig; wir gew&ouml;hnlich<lb/> unser Gesichtsbild nicht als etwas
   in<lb/> uns empfinden wie etwa einen Schmerz im Auge<lb/> da&szlig; wir aber wenn wir
   philosophieren<lb/> geneigt sind diesem <del type="dnpc">Vergleich</del> Bild
   gem<corr type="trsn"><orig type="trsn1">a</orig><reg type="trsn2">&auml;</reg></corr>&szlig;<lb/> zu denken&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,34v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2"><del type="dnpc">&udq.sudq;</del>The
   &usq.susq;if<del type="dnpc">&usq.eusq;</del>&div;sensation&usq.eusq;<del type="dnpc">&udq.eudq;</del>&p.es;</s>
  
  <s type="es">Compare with the &usq.susq;table&div;sen<lb rend="shyphen"/>sation&usq.eusq;&p.es;</s> 
  <s type="es">There is the question &udq.sudq;<c type="c">W</c>hat<corr type="tra">&app.contr;</corr>s the
   table<lb/><corr type="tra">&divl;</corr>sensation like&udq.eudq; &amp.and; the answer is a
   picture of a<lb/> table&p.es;</s> 
  <s type="es">In what sense is the if<corr type="tra">&div;</corr>sensation<lb/> analogous to the
   table<corr type="tra">&div;</corr>sensation&qm.eis;</s> 
  <s type="es">Is there<lb/> a description of this sensation &amp.and; <emph rend="us1">what do we<lb/> call a
   description of it</emph>&p.eis;</s> 
  <s type="es">Putting the gestures in<lb rend="shyphen"/>stead of the sensation means
   <del type="dnpc">really</del> just<lb/> <del type="dnpc"><corr type="npcn">pr</corr></del> giving the nearest rough
   description there<lb/> is of <choice type="dsl"><orig type="alt1"><del type="d">this</del></orig>  <orig type="alt2"> <add rend="i">the</add></orig></choice>
   <c type="c">E</c>xperience<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="bl_2"/></ab>
  
 

<ab n="Ms-148,34v[4]et35r[1]et35v[1]et36r[1]et36v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm" wabmarks-ml="8" wabmarks-ml_duplikat="9">
  <s type="es" rend="indl_2">Example<lb rend="hl"/><emph rend="bl_1"/> &lb;&udq.sudq;I have a peculiar feeling of pastness in my
  wrist&udq.eudq;&p.es;&rb;</s><lb rend="hl"/> 
 <pb facs="Ms-148_35r" rend="recto" n="pagename_Ms-148,35r pageref_Ms-148,71"/><fw type="pageno">19&p.ordinal;</fw>
  <s type="es">6&rp; &udq.sudq;<c type="c">W</c>e shall never know whether he meant this<lb/> or
   that&udq.eudq;&p.es;</s> 
  <s type="es"><seg type="notation" ana="p" rend="literal"><choice type="o"><orig type="o1">B</orig><orig type="o2">C</orig></choice></seg> died after the training in that<lb/> room&p.es;</s> 
  <s type="es">We say&colon; &udq.sudq;<c type="c">P</c>erhaps he would have<lb/>
   rea<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>ted like <seg type="notation" ana="p" rend="literal">B</seg> when taken into the
   daylight&p.es;<corr type="tra">&udq.eudq;</corr></s> <lb/>
  <s type="es">But we shall never know&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_4">&alpha;&rp; We should say this question was decided<lb/> if he arose from his
   grave &amp.and; we then made<lb/> the experiment with him&p.es;</s> 
  <s type="es">Or his ghost appeared<lb/> to us in a spiritualist s'an<corr type="trsn"><orig type="trsn1">s</orig><reg type="trsn2">c</reg></corr>e
   &amp.and; told<lb/> us that he has a certain experience&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_3">&beta;&rp; We don<corr type="tra">&app.contr;</corr>t accept any evidence&p.es;</s><lb/> 
  <s type="es">But what if we didn&app.contr;t accept the evidence<lb/> in 5 either &amp.and;
   said <emph rend="uw1"><del type="dnpc"><corr type="npcn">somet</corr></del> <add rend="i">&lp;something like&rp;</add></emph>
   &udq.sudq;<c type="c">W</c>e can&app.contr;t<lb/> be sure that <choice type="o"><orig type="o1">it</orig><orig type="o2">he</orig></choice> is the
   identical man<lb/> who was trained in the room&udq.eudq;, or&colon; &udq.sudq;he
   is<lb/> the identical man but we can&app.contr;t know whether<lb/> he <emph rend="us1">would
   have</emph> behaved like this in the<lb/> past time when he was
   trained<corr type="tra">&udq.eudq;</corr>&p.es;</s><lb rend="hl"/> 
  <s type="es">7&rp; <c type="c">W</c>e introduce a <add rend="im">new</add> notation for the<lb/>
   expression<del type="dnpc">&colon;</del> &udq.sudq;<c type="c">I</c>f <seg type="notation" ana="p" rend="literal"><choice type="o"><orig type="o1">A</orig><orig type="o2">P</orig></choice></seg> happens
   then always &lp;as a rule&rp;<lb/> <add rend="our"><seg type="notation" ana="p" rend="literal"><choice type="o"><orig type="o1">B</orig><orig type="o2">Q</orig></choice></seg></add> happens&p.es;</s>
  
  <s type="es"><seg type="notation" ana="p" rend="literal">P</seg> didn<corr type="tra">&app.contr;</corr>t happen this time &amp.and;
   <del type="dnpc">so</del> <seg type="notation" ana="p" rend="literal">Q</seg> didn&app.contr;t<lb/> happen<corr type="tra">&p.es;</corr>&udq.eudq;</s> 
  <s type="es">We say instead&colon; &udq.sudq;<c type="c">I</c>f <seg type="notation" ana="p" rend="literal">P</seg> had happened<lb/> <seg type="notation" ana="p" rend="literal">Q</seg>
   would<del type="dn"><gap extent="characters_1"/></del> have happened&udq.eudq;&p.es;</s> 
  <s type="es"><abbr type="abb">E&p.abb;g&p.abb;</abbr> &udq.sudq;<c type="c">I</c>f the gunpowder<lb/> is dry
   <add rend="im">under these circumstances</add> a spark of this strength explodes
   it&p.es;</s> 
  <s type="es">It<lb/> wouldn&app.contr;t dry this time &amp.and; <add rend="i">under the same
   circumstances</add> didn&app.contr;t explode<corr type="tra">&p.es;</corr>&udq.eudq;</s> 
  <s type="es">We say<lb/> instead &udq.sudq;<c type="c">I</c>f the gunpowder had been dry this time<lb/> it
   would have exploded&udq.eudq;&p.es;</s> 
  <s type="es">The point of this<lb/> notation is that it nears the form of this<lb/> preposition
   very much to the form&colon; &udq.sudq;<c type="c">T</c>he <unclear>gunpowder</unclear>
   <pb facs="Ms-148_35v" rend="verso" n="pagename_Ms-148,35v pageref_Ms-148,72"/><fw type="pageno" place="top left">20&p.ordinal;</fw> was dry this time so it
   exploded&udq.eudq;<corr type="tra">&p.es;</corr></s> 
  <s type="es"><c type="k">I</c> mean<lb/> the new <unclear>form</unclear> doesn&app.contr;t stress the fact<lb/> that
   it did not explode but, we<lb/> might say, points a vivid picture of<lb/> it exploding
   this time&p.es;</s> 
  <s type="es">We could imagine<lb/> <del type="dnpc">a</del> <add rend="im">two forms of expression in a</add>
   picture language <del type="dnpc">which was</del> corresponding<lb/> to the two kinds of
   notations in the word<lb/> language&p.es;</s> <del type="dnpc">
  <s type="es">The second notation would<lb/> consist in actually painting a picture<lb/> of the
   explosion&p.es;</s> </del>
  <s type="es">The second notation<lb/> will be particularly appropriate
   <add rend="im"><abbr type="abb">e&p.abb;g&p.abb;</abbr></add> if<lb/> we wish to give a person a shock by<lb/>
   making him vividly imagine <choice type="dsl"><orig type="alt1"><del type="d">what</del></orig>  <orig type="alt2"> <add rend="i">that which</add></orig></choice>
   would<lb/> have ha<corr type="tran">p</corr>pened, stressing only slightly that<lb/> it
   <choice type="em"><orig type="em1"><del type="d">hasn&app.contr;t</del> <add rend="i">didn&app.contr;t</add>
   hap<corr type="tran">p</corr><del type="dn">p</del>en<del type="dn">ed</del></orig>  <orig type="em2"><choice type="dsl"><orig type="alt1"><del type="d">hasn&app.contr;t
   happened</del></orig>  <orig type="alt2"> didn&app.contr;t happen</orig></choice></orig></choice>&p.es;</s><lb rend="hl"/> <emph rend="indl_2"/><seg type="wabmarks-ml">8</seg>
  <s type="es">Someone might say to us&colon; &udq.sudq;<c type="c">B</c>ut are<lb/> you sure that the
   second sentence means<lb/> just what the first one means &amp.and; not
   <del type="dnpc"><gap extent="words_1"/></del><lb/> just something similar or <add rend="im">that &amp.and;</add> something
   else<lb/> as well&qm.eis;</s> 
  <s type="es">&lp;<persName key="Moore, George Edward" corresp="commentary">Moore</persName>&rp;</s> 
  <s type="es"><c type="k">I</c> should say&colon; <c type="k">I</c>&app.contr;m talking<lb/> of the case where it
   means just this,<lb/> <del type="dnpc">if it&app.contr;s used in</del> &amp.and; this seems
   to me<lb/> an important case &lp;which you <unclear>caused<corr type="npcn">e</corr></unclear><lb/> by
   saying what you have said&rp;&p.es;</s> 
  <s type="es">But<lb/> of course I don&app.contr;t say that it<lb/> isn&app.contr;t used in other
   ways as well &amp.and; then<lb/> we&app.contr;ll have to talk about these<lb/> other
   cases separately&p.es;</s><lb rend="hl"/> <emph rend="indl_2"/><seg type="wabmarks-ml">9</seg>
  <s type="es">Someone says &dash;&udq.sudq;lowering one<corr type="tra">&app;</corr>s voice some 
   <pb facs="Ms-148_36r" rend="recto" n="pagename_Ms-148,36r pageref_Ms-148,73"/><fw type="pageno">21</fw> times means
   
   
   <reloc type="fetch-nec" n="Ms-148,36r_Ms-148,36r" corresp="Ms-148#1" rend="endarrback startarrback">that
   what you say is less important than<lb/> the rest</reloc>
   <reloc type="fetch-nec" n="Ms-148,36r_Ms-148,36r" corresp="Ms-148#2" rend="endarrback startarrback">&amp.and; in other cases you lower your voice to show<lb/></reloc> that you
   w<corr type="npcn">h</corr>ish to draw<lb/> special attention to what you now say
   
   <reloc type="relocate-nec" n="Ms-148,36r_Ms-148,36r" corresp="Ms-148#2"><emph rend="startarrback"/>  in<lb/> other cases you
   lower your voice to show<lb/></reloc>
   <reloc type="relocate-nec" n="Ms-148,36r_Ms-148,36r" corresp="Ms-148#1"><emph rend="endarrback"/> that what you say is less
   important than<lb/> the rest</reloc>&p.es;<corr type="tra">&udq.eudq;</corr></s><lb rend="hl"/> 
  <s type="es" rend="indl_2"><choice type="s"><orig type="alt1">We</orig>  <orig type="alt2"> <add rend="i"><c type="c">I</c>t</add></orig></choice> must be clear that our examples are<lb/> not
   preparations to the <del type="dnpc"><corr type="npcn"><gap extent="words_1"/></corr></del> analysis<lb/> of the actual meaning
   of the expression<lb/> so &amp.and; so <del type="dnpc"><corr type="npcn">Aut</corr></del>
   &lp;<persName key="Nicod, Jean" corresp="commentary">Ni<corr type="trsn"><orig type="trsn1">qu</orig><reg type="trsn2">c</reg></corr>od</persName>&rp; but
   giving them effects<lb/> that &udq.sudq;analysis&udq.eudq;&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">11<del type="dnpc"><corr type="npcn">Hab</corr></del> <c type="c">H</c>ave we now shown that to<lb/> say <add rend="im">in
   5</add> &udq.sudq;<c type="c">W</c>e can&app.contr;t know whether he would have behaved<lb/>
   &sp.sms;&udq.eudq; makes no sense&qm.eis;</s> 
  <s type="es">We should say<lb/> <choice type="s"><orig type="alt1">the sentence</orig>  <orig type="alt2"> <add rend="i">to say <emph rend="uw1">this</emph> sentence</add></orig></choice>
   under these circumstances <add rend="im">has</add><lb/> lost <choice type="s"><orig type="alt1"><emph rend="uw1">its</emph></orig>  <orig type="alt2">
   <add rend="i"><emph rend="uw1">the</emph></add></orig></choice> point which it would have had<lb/> under other
   <abbr corresp="circumstances">circs&p.abb;</abbr> but this
   doesn<corr type="tra">&app.contr;</corr>t mean that we<lb/> can&app.contr;t give it another
   point&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">10&rp; <c type="c">W</c>e say &udq.sudq;<c type="c">W</c>e <choice type="dsl"><orig type="alt1"><del type="d">don&app.contr;t</del></orig>  <orig type="alt2">
   <add rend="i">can<corr type="tra">&app.contr;t</corr></add></orig></choice> know whether this spark<lb/> would have
   been sufficient to ignite that mixture;<lb/> because we can&app.contr;t reproduce
   the exact mixture<lb/> not having the exact ingredients or not having a<lb/> balance
   to weigh them <abbr type="abb">etc<corr type="tra">&p.abb;</corr></abbr>
   <abbr type="abb">etc&p.abb;</abbr><corr type="tra">&udq.eudq;</corr></s> 
  <s type="es">But suppose<lb/> we could reproduce all the circumstances &amp.and; someone<lb/>
   said &udq.sudq;we can&app.contr;t know whether it would have<lb/>
   exploded&udq.eudq; <choice type="s"><orig type="alt1">as we can&app.contr;t know whether</orig>  <orig type="alt2"> <add rend="i">&amp.and;
   being asked why he said because</add></orig></choice> <emph rend="us1">under these<lb/> circumstances</emph> it
   would have exploded <emph rend="us1">then</emph>&p.es;&udq.eudq;</s><lb/> 
  <s type="es">This answer would set our head whirling&p.es;</s> 
  <s type="es">We should<lb/> feel he wasn&app.contr;t playing the same game with<lb/> that
   expression as we do&p.es;</s> 
 <s type="es">We should be  <pb facs="Ms-148_36v" rend="verso" n="pagename_Ms-148,36v pageref_Ms-148,74"/><fw type="pageno" place="top left">22&p.ordinal;</fw>
   inclined to say &udq.sudq;<c type="c">T</c>his makes no sense&em.ees;&udq.eudq;</s><lb/> 
  <s type="es">And this means that we are at a loss<lb/> not knowing what reasoning, what
   actions<lb/> go with this expression&p.es;</s> 
  <s type="es">Moreover we believe<lb/> that he made up a sentence analogous<lb/> to sentences used
   in certain <abbr corresp="language">lang&p.abb;</abbr><lb/> games <del type="dnpc">but</del> not
   noticing that he<lb/> took the <emph rend="us1">point</emph> away<del type="dn">t</del>&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">In which case do we say that a<lb/> sentence has <corr type="tra">a</corr> point&qm.eis;</s> 
  <s type="es">That comes to<lb/> asking in which case do we call<lb/> something a language
   game&p.es;</s> 
  <s type="es"><c type="k">I</c><lb/> can only answer&p.es;</s> 
  <s type="es">Look at the family<lb/> of language games <corr type="tra">&amp.and;</corr> that will show<lb/>
   you whatever can be <emph rend="uw1">shown</emph> about<lb/> the matter&p.es;</s><lb rend="hl"/> <del type="dnpc">
  <s type="es" rend="indl_2">12&rp; &udq.sudq;<c type="c">W</c>e can never know what he really sees,<lb/> for he has
   his own visual image &amp.and; I have<lb/> mine&udq.eudq;&p.es;</s> 
  <s type="es">&amp.and; we can&app.contr;t say</s><lb rend="hl"/> </del></ab>
  
 

<ab n="Ms-148,36v[2]et37r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm"> 
  <s type="es" rend="indl_2">12&rp; &lp;<c type="c">T</c>he private visual image&p.es;&rp;</s> 
  <s type="es"><seg type="notation" ana="p" rend="literal">B</seg> is trained to<lb/> describe his afterimage when he has looked<lb/> say into
   a bright red light&p.es;</s> 
  <s type="es">He is made to<lb/> look into the light, &amp.and; then to shut his eyes<lb/>
   &amp.and; he is then asked &udq.sudq;<c type="c">W</c>hat do you
   see&qm.eis;&udq.eudq;&p.es;</s> 
  <s type="es">This<lb/> question before was put to him only if he<lb/> looked at physical
   objects&p.es;</s> 
  <s type="es">We suppose he<lb/> reacts by a description of what he sees with<lb/> closed
   eyes&p.es;&dash;</s> 
  <s type="es">But halt&em.ees;</s> 
 <s type="es">This description<lb/> of the training seems wrong for what if 
  <pb facs="Ms-148_37r" rend="recto" n="pagename_Ms-148,37r pageref_Ms-148,75"/><fw type="pageno">23</fw> <emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-37r.bmp" ana="graphics_Zeichen;  Gekritzel" rend="bitmap">148082</seg></emph> I had had to describe my
   own, not <seg type="notation" ana="p" rend="literal">B&app.contr;s</seg>, training&p.eis;</s><lb/> <add rend="iupm">
    <s type="es"><corr type="npc"><gap extent="words_1"/></corr></s> </add>
  <s type="es">Would I then also have said&colon; &udq.sudq;I reacted<lb/> to the question by
   &sp.sms;&udq.eudq; &amp.and; not rather&colon; &udq.sudq;<c type="c">W</c>hen<lb/> I had
   <del type="dnpc"><corr type="npcn">clo</corr></del> closed my eyes I saw an image &amp.and;<lb/> described
   it&udq.eudq;&p.es;</s> 
  <s type="es">If I say &udq.sudq;I saw an image<lb/> &amp.and; described it I say this as
   opposed to the<lb/> case <choice type="s"><orig type="alt1">where</orig>  <orig type="alt2"> <add rend="i">in which</add></orig></choice> I <del type="dnpc">described</del>
   gave a description without<lb/> seeing an image&p.es;</s> 
  <s type="es">&lp;<c type="k">I</c> might have lied or <emph rend="uw1">not</emph>&p.es;&rp;</s><lb/> 
  <s type="es">Now we could <add rend="im">of course</add> also distinguish these cases<lb/> if <seg type="notation" ana="p" rend="literal">B</seg>
   describes an afterimage&p.es;</s> 
  <s type="es">But we don<corr type="tra">&app.contr;</corr>t wish<lb/> <del type="dnpc">to distinguish them but we
   only</del> <add rend="im">to</add> consider now<lb/> cases in which the mechanism of lying
   plays any<lb/> part&p.es;</s> 
  <s type="es">For <choice type="o"><orig type="o1">I</orig><orig type="o2">i</orig></choice>f you say &udq.sudq;I always know whether<lb/> I am lying
   but not whether the other <del type="dnpc"><corr type="npcn">ma</corr></del><lb/> person is&udq.eudq;, I
   say&colon; in the case I&app.contr;m considering<lb/> I can&app.contr;t be said
   to <emph rend="us1">know</emph> that I&app.contr;m not<lb/> lying, or let us say not saying the
   <add rend="im">un</add>truth,<lb/> because the dilem<corr type="tran">m</corr>a saying the truth or the<lb/>
   untruth is in this case unknown to me&p.es;</s> 
  <s type="es"><choice type="dsl"><orig type="alt1"><del type="d">Think of the fact</del></orig>  <orig type="alt2"> <add rend="i"><c type="c">R</c>emember</add></orig></choice> that when
   I&app.contr;m asked<lb/> &udq.sudq;what do you see here&udq.eudq; I
   don&app.contr;t always ask<lb/> myself&colon; &udq.sudq;<c type="c">N</c>ow shall I say the
   truth or <del type="dnpc">not</del><lb/> something else&qm.eis;&udq.eudq;</s> 
  <s type="es">If you say &udq.sudq;but surely if<lb/> you <add rend="im">in fact</add> speak the truth
   then you did see some<lb rend="shyphen"/>thing &amp.and; you saw what you said you
   saw&udq.eudq;<corr type="tra">&p.es;</corr></s><lb/> 
  <s type="es"><c type="k">I</c> answer&colon; <c type="c">H</c>ow can I know that I see what I<lb/> say I
   see&qm.eis;</s> 
  <s type="es">Do I have a criter<corr type="trsn"><orig type="trsn1">ion</orig><reg type="trsn2">ium</reg></corr> one
   <del type="dnpc"><corr type="npcn">tha</corr></del> for the <unclear>colour</unclear> I see actually being
   red&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/> 
 <pb facs="Ms-148_37v" rend="verso" n="pagename_Ms-148,37v pageref_Ms-148,76"/><fw type="pageno" place="top left">24&p.ordinal;</fw></ab>
    
 

<ab n="Ms-148,37v[1]et38r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 
 <emph rend="indl_2"/><emph rend="dlilm">
  <s type="es">13&rp; <add rend="im"><c type="c">W</c>e imagine that</add> <corr type="trsn"><orig type="trsn1">T</orig><reg type="trsn2">t</reg></corr>he expression
   &udq.sudq;I can&app.contr;t see what you see&udq.eudq;<lb/> has been given sense
   by explaining it<lb/> to mean&colon; &udq.sudq;<c type="k">I</c> can&app.contr;t see what
   you see being<lb/> in a different position relative to the object<lb/> we are looking
   at&udq.eudq;, or &udq.sudq;&sp.sms; having not as<lb/> good eyes as
   you&udq.eudq;, or &udq.sudq;&sp.sms; having found<lb/> as in &sp.sms; that
   <seg type="notation" ana="p" rend="literal">B</seg> sees something which<lb/> we don&app.contr;t thought we look at the same
   <c type="c">O</c>bject&p.es;<corr type="tra">&udq.eudq;</corr></s><lb/> 
  <s type="es"><abbr type="abb">etc&p.abb;</abbr></s> 
  <s type="es"><c type="k">I</c> can&app.contr;t see your afterimage might<lb/> be explained to mean I
   can&app.contr;t see what<lb/> you see if I close my eyes meaning<lb/> <del type="dnpc">when</del>
   you say you see a red circle<corr type="tra">,</corr> I see<lb/> a yellow one&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">14&rp; <c type="c">I</c>dentity of physical objects, of shapes,<lb/> colours, dreams,
   toothache&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">15&rp; &lp;<c type="c">T</c>he <choice type="dsl"><orig type="alt1"><del type="d">thing</del></orig>  <orig type="alt2"> <add rend="i">object</add></orig></choice> we see&rp;
   <c type="c">T</c>he physical <c type="c">O</c>bject &amp.and;<lb/> its appearance&p.es;</s> 
  <s type="es">Form of expression&colon; different<lb/> views of the same
   <abbr corresp="physical">phys&p.abb;</abbr> object are different<lb/> objects
   seen&p.es;</s> 
  <s type="es">We ask &udq.sudq;<c type="c">W</c>hat do you see&udq.eudq;<lb/> &amp.and; he can either
   answer &udq.sudq;a chair&udq.eudq;, or<lb/> &ldq.sldq;this&udq.eudq;
   &lp;&amp.and; draw the particular view of the<lb/> chair&rp;&p.es;</s> 
  <s type="es">So we are now inclined to say<lb/> that each man sees a different object<lb/>
   &amp.and; one which no other person sees, for<lb/> even if they look at the same
   chair from<lb/> the same spot it may appear different<lb/> to them &amp.and; the
   objects before the other<lb/> minds eye I can&app.contr;t look at&p.es;</s> 
  </emph><lb rend="hl"/> 
  <s type="es" rend="indl_2"><emph rend="dlilm">16&rp; &lp;<c type="k">I</c> can&app.contr;t know whether he sees
   anything</emph></s> <pb facs="Ms-148_38r" rend="recto" n="pagename_Ms-148,38r pageref_Ms-148,77"/><fw type="pageno">25</fw> at all or
  only behaves as I do when I see<lb/> something&p.es;&rp;
  <s type="es">There seems to be an undoubted<lb/> asym<corr type="tran">m</corr>etry in the use of the
   word &udq.sudq;<choice type="dsl"><orig type="alt1"><del type="d">I</del> </orig>  <orig type="alt2"><add rend="i">to</add></orig></choice> see&udq.eudq; &lp;&amp.and;
   all<lb/> words relating to personal experience&rp;&p.es;</s> 
  <s type="es">One<lb/> <choice type="s"><orig type="alt1">can</orig>  <orig type="alt2"> <add rend="i">is inclined to</add></orig></choice> state this in the way that
   &udq.sudq;I know when<lb/> I see something by just seeing it, without hearing<lb/>
   what I say or observing the rest of my behaviour<lb/> whereas I know
   <emph rend="us1">that</emph> he sees &amp.and; <emph rend="us1">what</emph> he sees<lb/> only by observing
   his behaviour, <abbr type="abb">i&p.abb;e&p.abb;</abbr> indirectly&udq.eudq;&p.es;</s><lb rend="hl"/> 
  
  <s type="es" rend="indl_2">a&rp; <c type="c">T</c>here is a mistake in this <gap extent="words_1"/>&colon; <c type="k">I</c> know
   what<lb/> I see because I see it&udq.eudq;&p.es;</s> 
  <s type="es">What does it mean to<lb/> know that&p.eis;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2"><seg type="notation" ana="p" rend="literal">b</seg>&rp; <c type="c">I</c>t is true<del type="dn">t</del> to say that my reason for saying<lb/>
   that I see is not the observation of my<lb/> behaviour&p.es;</s> 
  <s type="es">But this is a <abbr corresp="grammatical  proposition">gramm&p.abb;prop&p.abb;</abbr><corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
  <s type="es" rend="indl_2"><seg type="notation" ana="p" rend="literal">c</seg>&rp; <c type="c">I</c>t seems to be an imperfection that I<lb/> can only know
   &sdash;&p.es;</s> 
  <s type="es">But this is just<lb/> the way we use the word &sdash;&p.es;&dash;</s> 
  <s type="es">Could we then<lb/> <del type="dnpc"/> &sp.sms; if we could&qm.eis;</s> 
 <s type="es">Certainly&p.es;</s><lb rend="hl"/>  <emph rend="bl_1"/> </ab>

 

<ab n="Ms-148,38r[2]et38v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2"><emph rend="slilm"><choice type="em"><orig type="em1">Does the person <add rend="i"><c type="c">S</c>hould we say that the
   person&sp.sms;</add> who has not learnt <add rend="i">the</add> language<lb/>
   know<del type="dn"><gap extent="characters_1"/></del><add rend="i">s</add></orig>  <orig type="em2"><choice type="s"><orig type="alt1"><c type="c">D</c>oes the person who has not learnt
   language<lb/> know</orig>  <orig type="alt2"> <c type="c">S</c>hould we say that the person who has not learnt the
   language<lb/> knows</orig></choice></orig></choice> <del type="dnpc"><corr type="npcn">wh</corr></del> that he sees red but
   can&app.contr;t express</emph><lb/> it&qm.eis;&dash;</s> 
  <s type="es">Or should we say&colon; &udq.sudq;he knows what he sees<lb/> but
   can&app.contr;t express it&udq.eudq;&qm.eis;&dash;</s> 
  <s type="es">So<del type="dnpc">,</del> besides seeing it<del type="dnpc">,</del> he<lb/> also knows what he
   sees&qm.eis;</s><lb rend="hl"/> 
  <s type="es" rend="indl_6">Imagine we described a totally different<lb/> experiment; say this, that I
   sting someone with<lb/> a needle &amp.and; observe whet<corr type="tran">h</corr>er he
   <choice type="s"><orig type="alt1">cries out or not</orig>  <orig type="alt2"> <add rend="i">makes a sound or not</add></orig></choice>&p.es;</s> 
 <s type="es">Then<lb/> surely it would interest us if the subject <del type="dnpc"><gap extent="words_1"/></del> 
  <pb facs="Ms-148_38v" rend="verso" n="pagename_Ms-148,38v pageref_Ms-148,78"/><fw type="pageno" place="top left">26</fw><choice type="s"><orig type="alt1">whenever we</orig>  <orig type="alt2"> <add rend="i">often when we</add></orig></choice>
   stung him saw, say, a red <choice type="o"><orig type="o1"/><orig type="o2">c</orig></choice>ircle&p.es;</s> 
  <s type="es">And<lb/> we would distinguish the case when he<lb/> cried out &amp.and; saw a circle
   from the case<lb/> when he cried out &amp.and; didn&app.contr;t see one&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_5">This case is quite straightforward &amp.and; there <choice type="s"><orig type="alt1">is<lb/> no problem
   about it&p.es;</orig>  <orig type="alt2"> <add rend="i">seems<lb/> to be nothing problematic in it&p.es;</add></orig></choice></s> 
 <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,38v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">If I say &udq.sudq;I tell myself that I<lb/> see red, I tell myself what I
   see&udq.eudq;<lb/> it seems that after having told<lb/> myself I now know better what
   I<lb/> see, am better acquainted with it,<lb/> than before&p.es;</s> 
 <s type="es">&lp;Now in a sense this may<lb/> be so &sp.ses;&rp;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,38v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">&udq.sudq;When he <del type="dnpc">said</del> asked me what <add rend="our">colours</add> I<lb/> saw,
   I guessed what he <choice type="s"><orig type="alt1"><emph rend="uw1">meant</emph></orig>  <orig type="alt2"> <add rend="i"><choice type="dsl"><orig type="alt1"><del type="d">wished</del></orig>  <orig type="alt2">
   wanted</orig></choice> to know</add></orig></choice> &amp.and; told<lb/> him&p.es;&udq.eudq;<del type="dnpc"><gap extent="words_1"/></del></s><lb rend="hl"/> 
 <emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,38v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">&udq.sudq;It is not enough to distinguish<lb/> between the cases in which
   <seg type="notation" ana="p" rend="literal">B</seg> <add rend="i">or I</add> sa<corr type="trsn"><orig type="trsn1">i</orig><reg type="trsn2">y</reg></corr>s that<lb/> I see red &amp.and; do see
   red &amp.and; the case <del type="dnpc">in which</del><lb/> I say this but don&app.contr;t
   see red; but we<lb/> must distinguish between the cases in<lb/> which <del type="dnpc">I say I
   see red,</del> <corr type="tra">I</corr> see red, say I<lb/> see red &amp.and; mean to describe
   what I see<lb/> &amp.and; the cases in which <del type="dnpc"/> I don&app.contr;t mean
   this&p.es;</s><lb rend="hl"/> <emph rend="bl_2"/>
 <pb facs="Ms-148_39r" rend="recto" n="pagename_Ms-148,39r pageref_Ms-148,79"/><fw type="pageno">27</fw></ab>
    
 

<ab n="Ms-148,39r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 
  <s type="es" rend="indl_2">Consider the case in which I don<corr type="tra">&app.contr;</corr>t<lb/> say what I see
   <add rend="i">in words</add> but by pointing to a<lb/> sample&p.es;</s> 
  <s type="es">Here again I distinguish now<lb/> between the cases in which I &usq.susq;just
   react<lb/> by pointing&usq.eusq; &amp.and; the case in which<lb/> I <emph rend="us1">see</emph>
   &amp.and; point&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/> </ab>
 
 

<ab n="Ms-148,39r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">Now suppose I asked&colon; &udq.sudq;<unclear>how</unclear> do<lb/> I know that I see
   <add rend="i">&amp.and; that I see red</add>&qm.eis;</s> 
  <s type="es">&udq.sudq;<abbr type="abb"><choice type="o"><orig type="o1">i</orig><orig type="o2"><c type="c">I</c></orig></choice>&p.abb;e&p.abb;</abbr> how do I know<lb/>
   that <add rend="i">I do</add> what you call seeing <add rend="im">&lp;&amp.and; seeing
   red&rp;</add>&qm.eis;&udq.eudq;</s> 
  <s type="es">For we use<lb/> the word &usq.susq;seeing&usq.eusq; &amp.and;
   &usq.susq;red&usq.eusq; <choice type="s"><orig type="alt1">between us&p.es;</orig>  <orig type="alt2"> <add rend="i">in a game we</add><lb/> play
    with one another&p.es;</orig></choice></s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,39r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">Don&app.contr;t you say&colon; &udq.sudq;<del type="dnpc"><c type="c">W</c>e not</del> <c type="c">I</c>n
   order to<lb/> be a description of our personal experi<lb rend="shyphen"/>ence
   <choice type="dsl"><orig type="alt1"><del type="d">it</del></orig>  <orig type="alt2"> <add rend="im"><emph rend="us1">what we say</emph></add></orig></choice> must not just be
   <choice type="dsl"><orig type="alt1"><del type="d">the</del></orig>  <orig type="alt2"> <add rend="i">our</add></orig></choice> reaction<lb/> but <add rend="im">must be</add>
   <emph rend="us1">justified</emph>&udq.eudq;&qm.eis; <del type="dnpc">by what I
   <unclear>see</unclear>&qm.eis;</del></s> 
  <s type="es">But<lb/> does the justification need another justifi<lb rend="shyphen"/>cation&qm.eis;</s><lb rend="hl"/> 
 <emph rend="bl_1"/></ab>
   

<ab n="Ms-148,39r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
  <s type="es" rend="indl_2"><del type="dnpc"><seg xml:lang="german">Nehmen wir</seg></del></s> <seg xml:lang="english"><s type="es"><c type="c">S</c>uppose<del type="dnpc"/>,
  <c type="c">W</c>e play the game 2<lb/> &amp.and; <seg type="notation" ana="p" rend="literal">B</seg> <del type="dnpc">tells</del> calls out the
  word &udq.sudq;red&udq.eudq;&p.es;</s>
 
 <s type="es">Suppose<lb/> <seg type="notation" ana="p" rend="literal">A</seg> <add rend="i">now</add> asks <seg type="notation" ana="p" rend="literal">B</seg>&colon; &udq.sudq;do you only say
  &usq.susq;red&usq.eusq; or did<lb/> you really see it&qm.eis;&udq.eudq;&p.es;</s></seg><lb rend="hl"/><emph rend="bl_1"/></ab>
 

<ab n="Ms-148,39r[5]et39v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;Surely there are two phenomena&colon; one, <add rend="i">just</add> speaking,<lb/>
   the other, seeing &amp.and; speaking accordingly&p.es;</s><lb/> 
  <s type="es">Answer&colon; <c type="c">C</c>ertainly we speak of these two cases<lb/> but we shall
   here have <corr type="trsn"><orig type="trsn1">s</orig><reg type="trsn2">t</reg></corr>o show how 
   <pb facs="Ms-148_39v" rend="verso" n="pagename_Ms-148,39v pageref_Ms-148,80"/><fw type="pageno" place="top left">28</fw> these expressions are used; or, in other<lb/>
   words, how they are taught&p.es;</s> 
  <s type="es">For the<lb/> mere fact<del type="dn">s</del> that we pos<corr type="tran">s</corr>ess a<lb/> picture of them
   does not help us<lb/> as we must describe <choice type="dsl"><orig type="alt1"><del type="d">how</del></orig>  <orig type="alt2"> <add rend="i">in what
   way</add></orig></choice> this picture<lb/> is used&p.es;</s> 
  <s type="es">More especially as we are<lb/> inclined to assure a use difference<lb/> from the
   actual one&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_4">We have therefore to explain under<lb/> what conditions we say&colon;
   &udq.sudq;I see &usq.susq;red&usq.eusq; but<lb/> don&app.contr;t see
   red&udq.eudq; or &udq.sudq;I saw &usq.susq;red&usq.eusq; &amp.and; see
   red&udq.eudq;, or<lb/> &udq.sudq;I <del type="dnpc">say <add rend="i"><gap extent="words_1"/></add> red</del> said
   &usq.susq;red&usq.eusq; but didn&app.contr;t see red&udq.eudq;
   <abbr type="abb">etc&p.abb;</abbr> <abbr type="abb">etc&p.abb;</abbr>&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_4">Imagine that saying red was often <del type="dnpc">are</del><lb/> followed by some
   agre<corr type="tran">e</corr>able event&p.es;</s> 
  <s type="es">We found<lb/> that the child enjoyed that event &amp.and; often<lb/> instead of
   &usq.susq;<add rend="our">green</add>&usq.eusq; said
   &usq.susq;<add rend="our">red</add>&usq.eusq;&p.es;</s> 
  <s type="es">We would use<lb/> this rea<choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>tion to play another
   <abbr corresp="language">lang&p.abb;</abbr><lb/> game with the child&p.es;</s> 
  <s type="es">We would say<lb/> &udq.sudq;you cheat, it<corr type="tra">&app;</corr>s
   red&udq.eudq;<corr type="tra">&p.es;</corr></s> 
  <s type="es">Now again we are<lb/> dependent upon the subsequent reaction of the
   child&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">Such games are actually played<lb/> with children&colon; <c type="c">T</c>elling a person
   the untruth<lb/> &amp.and; enjoying his surprise at finding out<lb/> what really
   happened&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,39v[2]et40r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_5">But couldn&app.contr;t we imagine some kind<lb/> of perversity in a child which
   made it<lb/> say red when it saw green &amp.and; <abbr corresp="vice versa">v&amp.and;
    v&amp.and;</abbr> &amp.and; at the <pb facs="Ms-148_40r" rend="recto" n="pagename_Ms-148,40r pageref_Ms-148,81"/><fw type="pageno">29</fw> same time this not being discovered because<lb/>
   it happened to see red <add rend="im">in these cases</add> when we say green&qm.eis;</s><lb/> 
  <s type="es">But if here we talk of perversity we<lb/> <choice type="s"><orig type="alt1">could</orig>  <orig type="alt2"> <add rend="i">might</add></orig></choice> also
   assure that we all were<lb/> perverse&p.es;</s> 
  <s type="es">For how are we or <seg type="notation" ana="p" rend="literal">B</seg> even to<lb/> find out that he is perverse&qm.eis;</s><lb rend="hl"/> 
  
  <s type="es" rend="indl_4">The idea is, that he <corr type="npc"><add rend="i"/></corr> finds out &lp;&amp.and; we do&rp;
   when<lb/> later on he le<corr type="tran">a</corr>rns how the word
   &usq.susq;perverse&usq.eusq; is<lb/> used &amp.and; <choice type="dsl"><orig type="alt1"><del type="d">now</del></orig>  <orig type="alt2">
   <add rend="i">then</add></orig></choice> he remembers that he was<lb/> that way all along&p.es;</s><lb rend="hl"/> 
  
  <s type="es" rend="indl_5">Imagine this case&colon; <c type="c">T</c>he child looks<lb/> at the lights&colon;
   sa<corr type="trsn"><orig type="trsn1">i</orig><reg type="trsn2">y</reg></corr>s the name of the<del type="dn">n</del> right<lb/> colour to himself
   in an a<corr type="npcn">s</corr>side &amp.and; then loud<lb/> the wrong word&p.es;</s> 
  <s type="es">It chuckles while doing<lb/> so&p.es;</s> 
  <s type="es">This is, one may say, a <emph rend="uw1">rudimentary</emph><lb/> form of cheating&p.es;</s> 
  <s type="es">One might even say&colon;<lb/> &udq.sudq;<c type="c">T</c>his child is going to be a
   liar&udq.eudq;&p.es;</s> 
  <s type="es">But<lb/> if it had not said the a<corr type="npcn">s</corr>side but<lb/> only imagined itself
   pointing to one colour<lb/> on the chart &amp.and; then said the wrong
   word,&dash;<lb/> was this cheating too&qm.eis;</s><lb rend="hl"/> 
  <s type="es" rend="indl_5">Can a child cheat like a banker without the knowledge of the
   banker&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,40r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> can assure you that before when I<lb/> said &usq.susq;I see
   red&usq.eusq; I saw black<corr type="tra">&p.es;&udq.eudq;</corr></s><lb rend="hl"/> <emph rend="bl_3"/> 
 <pb facs="Ms-148_40v" rend="verso" n="pagename_Ms-148,40v pageref_Ms-148,82"/><fw type="pageno" place="top left">30</fw></ab>
    
    
 

<ab n="Ms-148,40v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
  <s type="es" rend="indl_4">&udq.sudq;He tells us his private experience,<lb/> that experience which nobody
   but he<lb/> knows anything about&udq.eudq;&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,40v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;Surely his memory is worth more than<lb/> our
   dire<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>t criteria, as only he could<lb/> know what he
   saw&p.es;&udq.eudq;</s> <lb rend="hl"/><lb rend="hl"/></ab>
  
 

<ab n="Ms-148,40v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es"><corr type="npc"><gap extent="words_1"/></corr> <emph rend="indl_2"/><c type="c">B</c>ut let us see;&dash; <choice type="o"><orig type="o1">w</orig><orig type="o2"><c type="c">W</c></orig></choice>e
   sometimes say<lb/> ou<corr type="tran">t</corr>side philosophy such things as<lb/> &udq.sudq;of
   <emph rend="uw1">course</emph> only he knows how he feels&udq.eudq; <add rend="i">or &udq.sudq;I
   can&app.contr;t know what you feel&udq.eudq;</add>&p.es;</s><lb/> 
  <s type="es">Now how do we apply such a state<lb rend="shyphen"/>ment&qm.eis;</s> 
  <s type="es">Mostly it is an expression of<lb/> helplessness like &udq.sudq;I
   don&app.contr;t know what<lb/> to do&udq.eudq;&p.es;</s> 
  <s type="es">But this helplessness is not <add rend="our">due</add> to an unfortunate metaphysical<lb/>
   <del type="dnpc"><gap extent="words_1"/></del> fact, &usq.susq;the privacy of personal
   ex<lb rend="shyphen"/>perience&usq.eusq;, or it would worry us <del type="dnpc">the</del>
   <choice type="s"><orig type="alt1">always</orig>  <orig type="alt2"> <add rend="i">constantly</add></orig></choice>&p.es;</s><lb/> 
  <s type="es">Our expression is comparable to this&colon;<lb/>
   &udq.sudq;<c type="c">W</c>hat&app.contr;s done can&app.contr;t be
   undone&em.ees;&udq.eudq;&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,40v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
  <s type="es" rend="indl_2">We also say <add rend="im">to the <c type="c">D</c>octor</add> &udq.sudq;<c type="c">S</c>urely I must know
   whether<lb/> I have pains or not&em.ees;&udq.eudq;</s> 
 <s type="es">How do we use<lb/> this statement&qm.eis;</s><lb rend="hl"/><emph rend="bl_2"/> </ab>
  
 

<ab n="Ms-148,40v[5]et41r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;<corr type="trs"><orig type="trs1">Allright</orig> <reg type="trs2"><c type="c">A</c>ll right</reg></corr> if we can&app.contr;t
   talk in this way<lb/> about someone else I can certainly<lb/> say of myself that
   <emph rend="us1">I</emph> either saw red 
   <pb facs="Ms-148_41r" rend="recto" n="pagename_Ms-148,41r pageref_Ms-148,83"/><fw type="pageno">31</fw>
   <del type="dnpc"><gap extent="words_1"/></del> at that time or <choice type="s"><orig type="alt1">didn&app.contr;t</orig>  <orig type="alt2"> <add rend="i">had some other
   experience<corr type="npcn">&udq.eudq;</corr></add></orig></choice>&p.es;</s> 
  <s type="es"><c type="k">I</c> may<lb/> not remember now, but at the<lb/> time I saw one thing or the
   other&em.ees;&udq.eudq;</s><lb/> 
  <s type="es">This is like saying &udq.sudq;one of these<lb/> two pictures must have
   fitted&udq.eudq;&p.es;</s> 
  <s type="es">And<lb/> my answer is not that <add rend="im">perhaps</add> neither of them<lb/> fits but that
   I&app.contr;m not yet clear<lb/> about what &usq.susq;fitting&usq.eusq; in this
   case means&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,41r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Now is it the same case <corr type="tra">or</corr> are these<lb/> different cases&colon;
   <c type="c">A</c> blind man sees<lb/> everything just as we do but he acts<lb/> as a blind man
   does &amp.and; on the other<lb/> hand he sees nothing &amp.and; acts as<lb/> a blind
   man does&p.es;</s> 
  <s type="es">At first sight we<lb/> should say&colon; here we have obviously two<lb/> clearly
   different cases a<corr type="tran">l</corr>though we<lb/> admit <add rend="im">that</add> we
   can&app.contr;t know which we have<lb/> before us&p.es;</s> 
  <s type="es"><c type="k">I</c> should say&colon; <del type="dnpc"><c type="c">Y</c>ou</del> <c type="c">W</c>e obvious<lb rend="shyphen"/>ly
   use two different pictures which<lb/> <choice type="s"><orig type="alt1">one</orig>  <orig type="alt2"> <add rend="i">we</add></orig></choice> could describe like
   this&colon; &sp.ses;&p.es;</s> 
  <s type="es">But<lb/> we use <choice type="dsl"><orig type="alt1"><del type="d">both</del></orig>  <orig type="alt2"> <add rend="i">the</add></orig></choice> pictures in
   <add rend="our">such</add> <choice type="dsl"><orig type="alt1"><del type="d">same</del></orig>  <orig type="alt2"> <add rend="i">a</add></orig></choice> way<lb/> that the two games
   &usq.susq;come to the same&usq.eusq;&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/> </ab>
 
 

<ab n="Ms-148,41r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">By the way,&dash; would you say that he<lb/> <choice type="s"><orig type="alt1">surely</orig>  <orig type="alt2"> <add rend="i">certainly</add></orig></choice>
   knew that he was blind if he<lb/> was so&qm.eis;</s> 
 <s type="es">Why do you feel more reluctant<lb/> about this statement&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/> 
 <pb facs="Ms-148_41v" rend="verso" n="pagename_Ms-148,41v pageref_Ms-148,84"/><fw type="pageno" place="top left">32</fw></ab>
    
  
 

<ab n="Ms-148,41v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 
  <s type="es" rend="indl_4">&udq.sudq;Surely he knew that he saw red<lb/> but he
   <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>ouldn&app.contr;t say so&em.ees;&udq.eudq;&dash;</s> 
  <s type="es">Does that<lb/> mean &udq.sudq;<c type="c">S</c>urely he <choice type="em"><orig type="em1">saw <add rend="i">knew that</add>
   <seg type="notation" ana="p" rend="literal"><del type="d">t</del>he <choice type="o"><orig type="o1">co</orig><orig type="o2">sa</orig></choice>w</seg></orig>  <orig type="em2"><choice type="dsl"><orig type="alt1">saw</orig>  <orig type="alt2">knew that he saw</orig></choice></orig></choice>
   the colour<lb/> which we call &usq.susq;red&usq.eusq; &sp.sms;&udq.eudq;
   <del type="dnpc">&amp.and; if not</del>&dash; or<lb/> would you say it means &udq.sudq;he
   knew<lb/> that he saw <emph rend="us1">this</emph> colour&udq.eudq; &lp;pointing to a<lb/> red
   pa<add rend="el">t</add>ch&rp;<del type="dnpc">&udq.eudq;</del>&p.eis;</s> 
 <s type="es">But did he while he knew<lb/> it point to this patch&qm.eis;</s><lb rend="hl"/>  <emph rend="bl_1"/> </ab>

 

<ab n="Ms-148,41v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">Use of&colon; &udq.sudq;<c type="c">H</c>e knows what colour he
   sees&udq.eudq;&p.es;</s><lb/> 
  <s type="es">&udq.sudq;<c type="k">I</c> knew what colour I saw&udq.eudq;
   <abbr type="abb">etc&p.abb;</abbr><corr type="tra">&p.es;</corr></s> <lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,41v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german">
  <s type="es" rend="indl_2"><seg xml:lang="german">&udq.sudq;Nachdunkeln der Erinnerung&udq.eudq;</seg> <seg xml:lang="english">does
   this<lb/> expression make sense &amp.and; in what cases&p.es;</seg></s><lb rend="hl"/> 
 <s type="es">And isn&app.contr;t on<corr type="npcn">t</corr> the other hand the<lb/> picture which we use
  quite clear in<lb/> all cases&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,41v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">The case of old people usually <choice type="s"><orig type="alt1">having</orig>  <orig type="alt2"> <add rend="i">getting</add></orig></choice><lb/> memories of
   the time in which they learnt<lb/> to speak &amp.and; understand speech&colon;<lb rend="hl"/>
   <emph rend="indl_4"/> <seg type="notation" ana="p" rend="literal">a</seg>&rp; <c type="c">T</c>hey say or paint that such &amp.and;<lb/> such
   things have happened although<lb/> other records always contradict them<lb rend="hl"/>
   <emph rend="indl_3"/> <seg type="notation" ana="p" rend="literal">b</seg>&rp; <c type="c">T</c>he memories agree with the<lb/> records&p.es;</s>
  
  <s type="es"><choice type="o"><orig type="o1">I</orig><orig type="o2"><c type="c">O</c></orig></choice>nly in this case shall we<lb/> say that they remember
   <add rend="our">&sp.ses;</add>&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,41v[5]et42r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Suppose they paint the scenes they 
  <pb facs="Ms-148_42r" rend="recto" n="pagename_Ms-148,42r pageref_Ms-148,85"/><fw type="pageno">33</fw> say they remember &amp.and; paint the faces<lb/>
   very dark;&dash; shall we say that they<lb/> saw them that dark or that the<lb/>
  colour had become darker in their memo<lb rend="shyphen"/>ry&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,42r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">How do we know what colour a<lb/> person sees&qm.eis;</s> 
  <s type="es">By the sample he points<lb/> to&qm.eis;</s> 
  <s type="es">And how do we know <del type="dnpc"><corr type="npcn">wheth</corr></del><lb/> what relation the sample is
   <add rend="im">meant</add> to have<lb/> to the original&qm.eis;</s> 
  <s type="es">Now are we to say<lb/> &udq.sudq;we never know &sp.ses;&udq.eudq;&qm.eis;</s> 
  <s type="es">Or had we better<lb/> cut <del type="dnpc">out</del> these &udq.sudq;we never know
   &sp.sms;&udq.eudq; <add rend="im">out</add> of our<lb/> language &amp.and;
   <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>onsider how as a matter<lb/> of fact we are wont to use the<lb/> word
   &udq.sudq;to know&udq.eudq;&qm.eis;</s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,42r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">What if someone asked&colon; &udq.sudq;<c type="c">H</c>ow do <choice type="dsl"><orig type="alt1"><del type="d">we</del></orig>  <orig type="alt2">
   <add rend="i">I</add></orig></choice> know<lb/> that what I call <add rend="im">seeing</add> red is <add rend="im">not</add> an
   <emph rend="us1">entirely</emph> different<lb/> experience every
   time<del type="dnpc">&qm.eis;&udq.eudq;&qm.eis;</del> &amp.and; that I am not deluded<lb/>
   into thinking that it is the same or nearly<lb/> the
   same&qm.eis;&udq.eudq;&qm.eis;</s> 
  <s type="es">Here again the answer &udq.sudq;I<lb/> can&app.contr;t know &amp.and; the
   subsequent removal of the question<corr type="tra">&udq.eudq;</corr>&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Is it even true that when I call a<lb/> colour &usq.susq;red&usq.eusq; I
   <choice type="s"><orig type="alt1">serve myself of memory&qm.eis;&qm.eis;</orig>  <orig type="alt2"> <add rend="i">make use of
    memory&qm.eis;&qm.eis;</add></orig></choice></s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42r[5]et42v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">To use the memory of what happened  
  <pb facs="Ms-148_42v" rend="verso" n="pagename_Ms-148,42v pageref_Ms-148,86"/><fw type="pageno" place="top left">34</fw> when we were taught language is
   <corr type="trsn"><orig type="trsn1">allright</orig><reg type="trsn2">all right</reg></corr><lb/> as long as we don&app.contr;t think
  that<lb/> this memory teaches us something<lb/> essentially private&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;A rod has one length or another<lb/> how ever we find it
   out&p.es;&udq.eudq;</s> 
 <s type="es">Here again<lb/> the picture <seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-42v.bmp" ana="graphics_F&auml;den;  ungeteilt" rend="bitmap">148083</seg></s><lb rend="hl"/> </ab>
  
 

<ab n="Ms-148,42v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;Though he <choice type="o"><orig type="o1">k</orig><orig type="o2">c</orig></choice>an&app.contr;t say what it is he<lb/> sees
   while he is learning <abbr type="abb">No</abbr> 1, he&app.contr;ll<lb/> tell us afterwards what
   he saw&p.es;</s><lb/> 
  <s type="es">We mix this case up with the one&colon;<lb/> &udq.sudq;<c type="c">W</c>hen his
   <unclear>gag</unclear> will have been removed<lb/> he&app.contr;ll tell us what he
   saw&udq.eudq;&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">What does it mean &usq.susq;to tell <add rend="our">someone</add> what<lb/>
   <choice type="o"><orig type="o1">we</orig><orig type="o2">one</orig></choice> sees&usq.eusq;&qm.eis;</s> 
  <s type="es">Or &lp;perhaps&rp;, &usq.susq;to show someone what one
   sees&usq.eusq;&qm.eis;</s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,42v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">When we say &udq.sudq;he&app.contr;ll tell us what<lb/> he
   saw<corr type="tra">&udq.eudq;</corr> we have an idea that<lb/> then we&app.contr;ll know
   what he <emph rend="us1">really</emph> saw<lb/> in a dire<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>t way
   &lp;&udq.sudq;at least if he isn&app.contr;t
   lying&udq.eudq;&rp;<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;He is in a better position to say what he<lb/> sees than we
   are&p.es;&udq.eudq;&dash;</s> 
 <s type="es">That depends&p.es;&dash;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,42v[7]et43r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">If we say &udq.sudq;he&app.contr;ll tell us what he saw&udq.eudq;, it<lb/> is
   as though he would now make a  
   <pb facs="Ms-148_43r" rend="recto" n="pagename_Ms-148,43r pageref_Ms-148,87"/><fw type="pageno">35</fw> use of
   language which we had never taught<lb/> him&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,43r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">It is as if now <del type="dnpc"><corr type="npcn">whe</corr></del> we got an <emph rend="us1">insight</emph><lb/> into
   something which before we had only<lb/> seen from the outside&p.es;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,43r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Inside &amp.and; outside&em.ees;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,43r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;Our <choice type="s"><orig type="alt1">teaching</orig>  <orig type="alt2"> <add rend="i">training</add></orig></choice> connects the word
   &usq.susq;red&usq.eusq;<lb/> &lp;or is meant to connect it&rp; with a
   parti<lb rend="shyphen"/>cular impression of his &lp;a private impression<lb/> an impression in
   him&rp;&p.es;</s> 
  <s type="es">He then communicates<lb/> this impression&dash; indirectly, of course&dash;<lb/>
   through the medium of speech&p.es;&udq.eudq;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,43r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Where is <choice type="dsl"><orig type="alt1"><del type="d">the</del></orig>  <orig type="alt2"> <add rend="i">our</add></orig></choice> idea of &udq.sudq;direct
   &amp.and; indirect com<lb rend="shyphen"/>munication<corr type="tra">&udq.eudq;</corr> taken
   from&qm.eis;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,43r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">How, if we said, <add rend="im">as we sometimes might be inclined</add>&colon;
   &udq.sudq;<c type="c">W</c>e can only hope that<lb/> this&dash; indirect way of
   communication really<lb/> succe<choice type="o"><orig type="o1">d</orig><orig type="o2">e</orig></choice>ds<corr type="tra">&udq.eudq;</corr>&p.eis;</s><lb rend="hl"/>
 <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,43r[7]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">We <add rend="im">so long</add> see the facts about the usage of our<lb/> words crookedly
   <choice type="s"><orig type="alt1">as</orig>  <orig type="alt2"> <add rend="i">so</add></orig></choice> long as we are still<lb/> tempted <del type="dnpc"/>
   <add rend="im">here</add> to talk of direct &amp.and; indirect&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,43r[8]et43v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">As long as you use the picture indirect&div;<lb/>direct in this case you
   can&app.contr;t trust yourself 
   <pb facs="Ms-148_43v" rend="verso" n="pagename_Ms-148,43v pageref_Ms-148,88"/><fw type="pageno" place="top left">36</fw> about
   judging the grammatical situation<lb/> rightly <emph rend="uw1">otherwise</emph>&p.es;</s> <lb rend="hl"/>
 <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,43v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">In telling what one sees <emph rend="uw1">something<lb/> like</emph> turning one&app.gen;s
   inside out&qm.eis;</s> 
  <s type="es">And learning<lb/> to say what one sees, learning to let<lb/> others see inside
   us&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,43v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;We teach him to make us see what he<lb/> sees&udq.eudq;&p.es;</s> 
  <s type="es">He seems in an indirect way to<lb/> show us <emph rend="us1">the object</emph> which he sees,
   the object<lb/> which is before his mind<corr type="tra">&app;</corr>s eye&p.es;</s> 
  <s type="es">&udq.sudq;We can&app.contr;t<lb/> look at it, it is in him&p.es;&udq.eudq;</s><lb rend="hl"/> 
 <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,43v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">The idea of the private <emph rend="us1">object</emph> of vision&p.es;</s> <lb/>
 <s type="es">A<corr type="tran">p</corr>pearance, sense<corr type="tra">&div;</corr>datum&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,43v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">The visual field&p.es;</s> 
  <s type="es">&lp;<choice type="o"><orig type="o1">n</orig><orig type="o2"><c type="c">N</c></orig></choice>ot to be confused with visual<lb/> space&p.es;&rp;</s><lb rend="hl"/>
 <emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,43v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Telling someone what one sees seems<lb/> like showing him, if indirectly, the
   object<lb/> which is before one<corr type="tra">&app.contr;</corr>s mind&app.contr;s
   eye&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,43v[7]et44r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">The idea of the object before one&app.contr;s mind&app;s eye<lb/> is
   <choice type="dsl"><orig type="alt1"><del type="d">absolutely bound</del></orig>  <orig type="alt2"> <add rend="i">&lp;firmly&rp; tied</add></orig></choice> up with
   the<lb/> <add rend="our">idea</add> <del type="dnpc">that</del> of a comparis<corr type="npcn">s</corr>on of such<lb/>
   objects in different persons <del type="dnpc">which</del><lb/> compared to which the
   comparison 
   <pb facs="Ms-148_44r" rend="recto" n="pagename_Ms-148,44r pageref_Ms-148,89"/><fw type="pageno">37</fw> really
   used is <add rend="im">an</add> indirect one&p.es;</s><lb rend="hl"/>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,44r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">Whence the idea of the privacy of sensedata&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,44r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_2" xml:lang="english">
  <s type="es" rend="indl_2"><corr type="tra">&udq.sudq;</corr>But do you really wish to say that they<lb/> are not
   private <del type="dnpc">&lp;</del> that one person can<lb/> see the picture before the
   other person<corr type="tra">&app.contr;</corr>s<lb/> eye&qm.eis;&udq.eudq;</s> <emph rend="bl_2"/></ab>
  
 

<ab n="Ms-148,44r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Surely <add rend="im">you wouldn&app.contr;t think that</add> <emph rend="us1">telling</emph> someone
   what one sees <choice type="s"><orig type="alt1">is</orig>  <orig type="alt2"> <add rend="i">could be</add></orig></choice> a<lb/> more direct way of
   communicating than<lb/> showing him by pointing to a sample&em.ees;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,44r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;He&app.contr;ll tell us later what it was he saw&udq.eudq;<lb/> means
   that we&app.contr;ll get to know in a &lp;compara<lb rend="shyphen"/>tively&rp; direct
   <add rend="im">&amp.and; a sure</add> way what he saw as opposed<lb/> to the guesses we could
   make before&p.es;</s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,44r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">We don&app.contr;t real<del type="dn">l</del>ize that the answer he<lb/> gives us now is
   only part of a game like <abbr type="abb">No</abbr> 1<lb/> only more complicated&p.es;</s> <lb rend="hl"/>
 <emph rend="bl_7"/> <pb facs="Ms-148_44v" rend="verso" n="pagename_Ms-148,44v pageref_Ms-148,90"/><fw type="pageno" place="top left">38</fw></ab>
    
   
 

<ab n="Ms-148,44v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 
  <s type="es" rend="indl_2">We don<corr type="tra">&app.contr;</corr>t deny that he can remember <choice type="em"><orig type="em1"><del type="d">a</del>
   <add rend="i">having</add><lb/> dream<add rend="el">t</add></orig>  <orig type="em2"><choice type="s"><orig type="alt1"><del type="d">a</del><lb/> dream</orig>  <orig type="alt2"> having<lb/> dreamt</orig></choice></orig></choice> so
   &amp.and; so before he was born&p.es;</s> 
  <s type="es">Denying<lb/> this to us would be like denying that<lb/> he can say he remembers
   having dreamt<lb/> so &amp.and; so before he was born&p.es;</s><lb rend="hl"/> 
  <s type="es" rend="indl_4"><abbr type="abb">I&p.abb;e&p.abb;</abbr> we don&app.contr;t deny that he can<lb/> make this
   move but we say that<lb/> the move alone or together with all<lb/> the
   sensations<corr type="tra">,</corr> feelings <abbr type="abb">etc<corr type="tra">&p.abb;</corr></abbr> he might<lb/> have
   while he is making it does not<lb/> tell us <choice type="s"><orig type="alt1">what game it is a move
   of&p.es;</orig>  <orig type="alt2"> <add rend="i">to what game the move belongs&p.es;</add></orig></choice></s><lb rend="hl"/> 
  <s type="es" rend="indl_4">We might <abbr type="abb">e&p.abb;g&p.abb;</abbr> never try to connect up<lb/> a statement of
   this sort with anything<lb/> past &lp;in an other sense&rp;&p.es;</s> 
  <s type="es">We might treat<lb/> it as an interesting phenomenon &amp.and;<lb/> possibly connect
   it up with the <add rend="i">persons</add> <unclear>writing</unclear><lb/> in a
   <persName key="Freud, Sigmund" corresp="commentary">Freudian</persName> way or on the other hand<lb/> we
   may look for some phenomena in the<lb/> brain of the embryo which might be<lb/> called
   dreams <abbr type="abb">etc&p.abb;</abbr> <abbr type="abb">etc&p.abb;</abbr>&p.es;</s> 
  <s type="es">Or we may just<lb/> say&colon; &udq.sudq;old people are liable to say such<lb/>
   things&udq.eudq; &amp.and; leave it at that&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,44v[2]et45r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">Suppose now someone remembered that<lb/> <del type="dnpc">the</del> yesterday he called
   red &usq.susq;green&usq.eusq; &amp.and; <seg type="ct">vice<lb/> versa</seg> but that this
   didn&app.contr;t <choice type="o"><orig type="o1">p</orig><orig type="o2">a</orig></choice>ppear as he<lb/> also saw green what today he
   sees red<lb/> &amp.and; <seg type="ct">vice versa</seg>&p.es;</s> 
  <s type="es">Now here is a case in which<lb/> we might be inclined to say that we
   <pb facs="Ms-148_45r" rend="recto" n="pagename_Ms-148,45r pageref_Ms-148,91"/><fw type="pageno">39</fw> learn from him today something about<lb/> the
   working of his mind yesterday, that<lb/> yesterday we judged by the outside while<lb/>
   today we are allowed to look <del type="dnpc"><add rend="im"><gap extent="words_1"/></add></del> at the<lb/> inside of what
   happened&p.es;</s> 
  <s type="es">It is as though<lb/> we looked back but now got a glance<lb/> at something that was
   <choice type="s"><orig type="alt1">closed to us</orig>  <orig type="alt2"> <add rend="i">covered up</add></orig></choice> yester<lb rend="shyphen"/>day&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">If I say what it is I see how do<lb/> I compare what I say with what I<lb/> see in
   order to know whether I say<lb/> the truth&qm.eis;</s><lb rend="hl"/> 
  <s type="es" rend="indl_2">Lying about what I see, you might say,<lb/> is knowing what I see &amp.and;
   saying something<lb/> else&p.es;</s> 
  <s type="es">Supposing I said it just consists<lb/> of saying to myself &usq.susq;this is
   red&usq.eusq; &amp.and; aloud<lb/> &usq.susq;this is
   green&usq.eusq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">Compare lying &amp.and; telling the truth in the<lb/> case of telling what
   colour you see<lb/> with the case of describing a picture which<lb/> you saw or
   telling the right number<lb/> of things you had to count&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">Collating what you say &amp.and; what you see&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/> </ab>
 
 

<ab n="Ms-148,45r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
 <s type="es" rend="indl_2">Is there always a collating&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/> 
 <pb facs="Ms-148_45v" rend="verso" n="pagename_Ms-148,45v pageref_Ms-148,92"/><fw type="pageno" place="top left">40</fw></ab>
    
   
 

<ab n="Ms-148,45v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
 
  <s type="es" rend="indl_2">Or could you call it giving a picture of<lb/> the colour I see if I say the
   word<lb/> <corr type="tra">&udq.sudq;</corr>red<corr type="tra">&udq.eudq;</corr>&qm.eis; unless it be a
   picture by it<corr type="npcn">&app;</corr>s<lb/> connection with a sample&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">B<add rend="our">u</add>t isn&app.contr;t it <add rend="im">giving</add> a picture if I point to<lb/> a
   sample&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;What I show <emph rend="us1">reveals</emph> what I see&udq.eudq;;&dash;<lb/> in
   what sense does it do that&qm.eis;</s> 
  <s type="es">The<lb/> idea is that now you can so to speak<lb/> look inside me&p.es;</s> 
  <s type="es">Whereas I only reveal<lb/> to you what I see in a game of<lb/> revealing &amp.and;
   hi<choice type="o"><orig type="o1">dd</orig><orig type="o2">di</orig></choice>ng which is altogether<lb/> pla<corr type="trsn"><orig type="trsn1">i</orig><reg type="trsn2">y</reg></corr>ed with
   signs of one category<lb/> <del type="dnpc"><gap extent="words_1"/></del>
   &udq.sudq;<choice type="o"><orig type="o1">d</orig><orig type="o2"><c type="c">D</c></orig></choice>ire<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>t&div;indirect&udq.eudq;&p.es;</s><lb rend="hl"/>
 <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,45v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">We are thinking of a game in which<lb/> there is an inside in the normal
   sense&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,45v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm"><emph rend="indl_2"/><emph rend="dlilm">
  <s type="es">We must get clear about how the<lb/> metaphor of revealing &lp;outside
   &amp.and; inside&rp;<lb/> is actually applied by us; otherwise<lb/> we shall be
   tempted to look for<lb/> an inside behind that which in our<lb/> metaphor is the
   inside&p.es;</s> </emph><lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,45v[6]et46r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">We are used to describing the case<lb/> by mean of a picture which say
   <pb facs="Ms-148_46r" rend="recto" n="pagename_Ms-148,46r pageref_Ms-148,93"/><fw type="pageno">41</fw> contains 3 steps&p.es;</s> 
  <s type="es">But <add rend="im">when we think about language</add> we <emph rend="uw1">forget</emph> how<lb/> this
   picture is actually applied in<lb/> practical cases&p.es;</s> 
  <s type="es">We then are often<lb/> tempted to apply it as it wasn&app.contr;t<lb/>
   <choice type="em"><orig type="em1"><seg type="co">meant or<del type="dn">r</del>iginally</seg></orig>  <orig type="em2"><choice type="s"><orig type="alt1">meant originally</orig>  <orig type="alt2">
   originally meant</orig></choice></orig></choice> &amp.and; <del type="dnpc">lack</del> are puzzled<lb/> about a third
   step in the facts&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,46r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">&udq.sudq;<c type="k">I</c> see a particular <choice type="s"><orig type="alt1"><emph rend="uw1">sensedatum</emph></orig>  <orig type="alt2">
   <add rend="i"><emph rend="uw1">image</emph></add></orig> <orig type="alt3"> <add rend="i">thing</add></orig></choice> &amp.and; say<lb/> a particular
   thing&udq.eudq;&p.es;</s> 
  <s type="es">This is <corr type="trsn"><orig type="trsn1">allright</orig><reg type="trsn2">all right</reg></corr><lb/> if I real<choice type="o"><orig type="o1">l</orig><orig type="o2">i</orig></choice>se
   the way in which I specify<lb/> what I see &amp.and; what I say&p.es;</s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,46r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;If he had learnt to show me &lp;or tell me&rp;<lb/> what he sees, he
   could now show me&p.es;&udq.eudq;</s><lb/> 
  <s type="es">Certainly,&dash; but what is it like to<lb/> show me what he sees&qm.eis;</s> 
  <s type="es">It is pointing<lb/> to something under particular circum<lb rend="shyphen"/>stances&p.es;</s> 
  <s type="es">Or is it something else &lp;don<corr type="tra">&app.contr;</corr>t<lb/> be misled by the
   idea of <unclear>indirectness</unclear>&rp;<corr type="tra">&p.es;</corr></s><lb rend="hl"/> 
  <s type="es" rend="indl_2">You compare it with such a statement<lb/> as&colon; &udq.sudq;if he had learnt
   to open<lb/> up <choice type="s"><orig type="alt1">he could now open up &amp.and; show<lb/> me what&app.contr;s
   inside</orig>  <orig type="alt2"> <add rend="i">I could now see what&app.contr;s inside</add></orig></choice>&p.es;</s> 
  <s type="es"><c type="k">I</c> say yes, but<lb/> remember what opening up in this<lb/> case is
   like&p.es;</s><lb rend="hl"/>  <emph rend="bl_1"/> </ab>

 

<ab n="Ms-148,46r[4]et46v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">But what about the criteri<corr type="trsn"><orig type="trsn1">um</orig><reg type="trsn2">on</reg></corr> <unclear>whether</unclear><lb/> there
   is anything inside or not&qm.eis;</s> 
 <s type="es">Here we<lb/> say &udq.sudq;I know that there is something 
  <pb facs="Ms-148_46v" rend="verso" n="pagename_Ms-148,46v pageref_Ms-148,94"/><fw type="pageno" place="top left">42</fw> inside in <emph rend="us1">my</emph> case&p.es;</s> 
  <s type="es"><choice type="s"><orig type="alt1">And this is how<lb/> I know of the &usq.susq;inside&usq.eusq; at all first<lb/>
   hand&udq.eudq;&p.es;</orig>  <orig type="alt2"> &dobslash.alt;<c type="c">A</c>nd this is how I have first<lb/>
   hand knowledge of the inside at<lb/> all&p.es;&udq.eudq;</orig> <orig type="alt3"> &dobslash.alt;
   <corr type="npcn">&udq.sudq;</corr><c type="c">T</c>his is how I know about an<lb/> inside &amp.and; am
   led to suppose is in the other person too&p.es;&udq.eudq;</orig></choice></s><lb rend="hl"/> 
  <s type="es"><emph rend="startarrback"/><add rend="iupm"><emph rend="endarrback"/><emph rend="arrback">Further we are not inclined to
   say that only hitherto we have not known the mind of an other person but
   that the idea of this knowled<corr type="tran">g</corr>e is bound up with the idea of
   myself&p.es;</emph></add></s>  <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,46v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">&udq.sudq;So if I say &usq.susq;he has toothache&usq.eusq; I am<lb/> supposing
   that he has what I<lb/> have if I have toothache&p.es;&udq.eudq;</s> 
  <s type="es">Suppose I<lb/> said&colon; &udq.sudq;<c type="c">I</c>f I say <emph rend="us1">&usq.susq;I
   suppose&usq.eusq;</emph> he has<lb/> toothache<corr type="tra">,</corr> I am supposing that he<lb/>
   has what I have if I have <del type="dnpc"/>tooth<lb rend="shyphen"/>ache&udq.eudq;,&dash; this
   would be like saying<lb/> &udq.sudq;<c type="c">I</c>f I say &usq.susq;this cushion is
   red&usq.eusq; I mean<lb/> that it has the same colour which<lb/> the sofa has if it is
   red&udq.eudq;&p.es;</s> 
  <s type="es">But<lb/> this wasn&app.contr;t what <choice type="s"><orig type="alt1">I intended</orig>  <orig type="alt2"> <add rend="i">was meant</add></orig></choice> to<lb/>
   say with the first sentence&p.es;</s> 
  <s type="es"><c type="k">I</c><lb/> wished to say that talking about<lb/> his toothache at all was based<lb/>
   upon a supposition, a supposition which<lb/> <add rend="im">by its very
   <choice type="dsl"><orig type="alt1"><del type="d">nature</del></orig>  <orig type="alt2"> essence</orig></choice></add> could not be
   ver<corr type="trsn"><orig type="trsn1">y</orig><reg type="trsn2">i</reg></corr>fied&p.es;</s><lb rend="hl"/>   <emph rend="bl_1"/></ab>

 

<ab n="Ms-148,46v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english" emph="slilm">
  <s type="es" rend="indl_2">But if you look closer you will see<lb/> that this <emph rend="uw1">is</emph> an entire
   misrepresenta<lb rend="shyphen"/>tion of the use of the word
   &udq.sudq;toothache&udq.eudq;<corr type="tra">&p.es;</corr></s><lb rend="hl"/><emph rend="bl_1"/>  
 <pb facs="Ms-148_47r" rend="recto" n="pagename_Ms-148,47r pageref_Ms-148,95"/><fw type="pageno">43</fw></ab>
    
   
 

<ab n="Ms-148,47r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
 <emph rend="bl_1"/>
 <emph rend="sepline"/>
 <s type="es" rend="indl_2">Can two people have the same<lb/> afterimage&qm.eis;</s><lb rend="hl"/><emph rend="bl_1"/> </ab> 

<ab n="Ms-148,47r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
  <s type="es" rend="indl_2"><del type="dnpc"><seg xml:lang="german"><corr type="npcn">Spr</corr></seg></del>
   <seg xml:lang="english"><c type="c">L</c>anguage<corr type="tra">&div;</corr>game &usq.susq;<c type="c">D</c>escription of
   <choice type="s"><orig type="alt1">imaginings</orig>  <orig type="alt2"> <add rend="i">the picture before one<corr type="tra">&app.contr;</corr>s
    mind&app.gen;s eye</add></orig></choice>&p.es;&usq.eusq;</seg></s><lb rend="hl"/> <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,47r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Can two <del type="dnpc">people</del> persons have the<lb/> same picture before their
   mind&app;s eye&p.es;</s><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,47r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">In which case would we say that<lb/> they had two images exactly alike but<lb/> not
   identical&qm.eis;</s><lb rend="hl"/> <emph rend="bl_1"/><emph rend="sepline"/></ab>
  
 

<ab n="Ms-148,47r[5]et47v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">The fact that two ideas seem here<lb/> inseparably bound up suggests to<lb/> us that
   we are dealing with one idea<lb/> only &amp.and; not with two &amp.and; that by a<lb/>
   queer trick our language suggests a<lb/> totally different
   stru<corr type="trsn"><orig type="trsn1">k</orig><reg type="trsn2">c</reg></corr>ture of grammar<lb/> than the one actually
   used&p.es;</s> 
  <s type="es">For<lb/> we have the sentence that only I<lb/> can know direct<corr type="npcn">e</corr>ly my
   experience &amp.and;<lb/> only indirect<corr type="npcn">e</corr>ly the experience of<lb/> the
   <add rend="our">other</add> person&p.es;</s> 
  <s type="es"><choice type="em"><orig type="em1"><del type="d">This</del><add rend="i">us</add></orig>  <orig type="em2"><choice type="s"><orig type="alt1"><c type="c">T</c>his</orig>  <orig type="alt2"> <c type="c">T</c>hus</orig></choice></orig></choice>
   <add rend="i">language</add> suggests 4 possible<lb/> com<choice type="o"><orig type="o1">p</orig><orig type="o2">b</orig></choice>inations but rules out
   2&p.es;</s> 
 <s type="es">It is<lb/> as though I had used the 4 letters 
  <pb facs="Ms-148_47v" rend="verso" n="pagename_Ms-148,47v pageref_Ms-148,96"/><fw type="pageno" place="top left">44</fw> <seg type="notation" ana="p" rend="literal">a b c d</seg> to denote two objects only<lb/> but
   by my notation somehow<lb/> suggesting <del type="dnpc">the <corr type="npcn">pre</corr></del> that I am
  tal<lb rend="shyphen"/>king of 4&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,47v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">It seems as though I w<choice type="o"><orig type="o1">a</orig><orig type="o2">i</orig></choice>shed to<lb/> say that <del type="dnpc">I</del> <add rend="im">to
   me</add> <persName key="Wittgenstein, Ludwig" corresp="commentary"><abbr corresp="Ludwig  Wittgenstein">L&p.abb;W&p.abb;</abbr></persName> something applied<lb/>
   which does not apply to other<lb/> people&p.es;</s> 
  <s type="es">That is, there seems to be an<lb/> a<corr type="npcn">s</corr>symmetry&p.es;</s><lb rend="hl"/> 
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-47v.bmp" ana="graphics_Technische Darstellung; geometrischer  Beweis" rend="bitmap">148084</seg></emph></s><lb rend="hl"/> 
  <s type="es"><c type="k">I</c> <emph rend="us1">express</emph> things
   a<corr type="npcn">s</corr>symmetri<choice type="o"><orig type="o1">l</orig><orig type="o2">c</orig></choice>ally &amp.and; could<lb/> express them
   symmetrically; only then one<lb/> would see what facts prompt us to<lb/> the
   a<corr type="npcn">s</corr>symmetrical expression&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,47v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2"><c type="k">I</c> do this by spreading the <add rend="im">use of the</add> word I<lb/> over all
   <unclear>hu<choice type="o"><orig type="o1">p</orig><orig type="o2">m</orig></choice>an</unclear> bodies as opposed to<lb/>
   <persName key="Wittgenstein, Ludwig" corresp="commentary"><abbr corresp="Ludwig  Wittgenstein">L&p.abb;W&p.abb;</abbr></persName>
   al<corr type="npcn">l</corr>one&p.es;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,47v[4]et48r[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2"><c type="k">I</c> want to <del type="dnpc">bring</del> describe a situation<lb/> in which I should
   not be tempted to<lb/> say that I assumed or believed that<lb/> the other had what I
   have&p.es;</s> 
  <s type="es">Or, in other<lb/> words <del type="dnpc">I</del> a situation in which we would<lb/> not of
   <emph rend="us1">my</emph> <emph rend="us1">cons<choice type="o"><orig type="o1">t</orig><orig type="o2">c</orig></choice>iousness</emph> &amp.and; <emph rend="us1">his
   conscious<lb rend="shyphen"/>ness</emph>&p.es;</s> 
 <s type="es">And in which the idea would not <pb facs="Ms-148_48r" rend="recto" n="pagename_Ms-148,48r pageref_Ms-148,97"/><fw type="pageno">45</fw> occur to
  us that we could only be<lb/> conscious of our own consciousness&p.es;</s><lb rend="hl"/> <emph rend="bl_1"/></ab>
  
 

<ab n="Ms-148,48r[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es" rend="indl_2">The idea of the ego inhabiting a<lb/> body to be abolished&p.es;</s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,48r[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">If what any consciousness &tick; spreads over<lb/> all human bodies then
   the<corr type="trsn"><orig type="trsn1">ir</orig><reg type="trsn2">re</reg></corr> won<corr type="tra">&app.contr;</corr>t be<lb/> any temptation to
   use the word &usq.susq;ego&usq.eusq;<corr type="tra">&p.es;</corr></s> <lb rend="hl"/> <emph rend="bl_1"/></ab>
 
 

<ab n="Ms-148,48r[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">Let&app.contr;s assume that hearing was<lb/> done by no organ of the body we
   know<lb/> of<corr type="tra">&p.es;</corr></s><lb rend="hl"/><emph rend="bl_1"/> </ab>
  
 

<ab n="Ms-148,48r[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">Let us imagine the following arrangement&colon;<lb rend="hl"/><emph rend="bl_4"/></s> 
  <s type="es" rend="indl_2">If it is absurd to say that I only<lb/> know that <emph rend="us1">I</emph> see
   <del type="dnpc">&amp.and;</del> but not that the<lb/> others do,&dash; isn&app.contr;t this
   at any rate less<lb/> absurd than to say the opposite&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/> </ab>

<ab n="Ms-148,48r[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="german"><emph rend="indl_2"/><seg xml:lang="german">
  <s type="es">Ist eine Philosophie undenkbar die das<lb/> diametrale Gegenteil des
   Solipsismus ist&qm.eis;</s> </seg><lb rend="hl"/> 
 <s type="es" rend="indl_8"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-48r.bmp" ana="graphics_Technische Darstellung; geometrischer  Beweis" rend="bitmap">148085</seg></emph></s><lb rend="hl"/> 
 <pb facs="Ms-148_48v" rend="verso" n="pagename_Ms-148,48v pageref_Ms-148,98"/><fw type="pageno" place="top left">46&p.ordinal;</fw></ab>


<ab n="Ms-148,48v[1]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german"><emph rend="indl_2"/><seg xml:lang="english">
 
  <s type="es">The idea of the constituent of a<lb/> fact <del type="dnpc">&amp.and;</del>&colon;
   &udq.sudq;<c type="c">I</c>s my person &lp;or a person&rp; a<lb/> constituent of the
   fact<del type="dnpc"/> that<lb/> I see or not&udq.eudq;&p.es;</s> 
  <s type="es">This expresses a question<lb/> concerning the symbolism just as<lb/> if it were a
   question about <del type="dnpc"><unclear>the</unclear></del> nature&p.es;</s> </seg><lb rend="hl"/><emph rend="bl_1"/></ab>

<ab n="Ms-148,48v[2]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="german">
  <s type="es" rend="indl_2">&udq.sudq;Es denkt&udq.eudq;&p.es;</s> 
  <s type="es">Ist dieser Satz wahr &amp.und; &udq.sudq;ich<lb/> denke&udq.eudq;
   falsch&qm.eis;</s> <lb rend="hl"/><emph rend="bl_1"/></ab>
 

<ab n="Ms-148,48v[3]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2"><abbr corresp="Language&div;game"><c type="c">L</c>ang&p.abb;game</abbr>&colon; I
   paint, for myself, what I<lb/> see&p.es;</s> 
  <s type="es">The picture doesn&app.contr;t contain <emph rend="us1">me</emph>&p.es;</s><lb rend="hl"/> 
 <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,48v[4]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
 <s type="es"><emph rend="centered"><seg type="notation" corresp="http://wab.aksis.uib.no/cost-a32_fax/bmp/148/notatio148-48v.bmp" ana="graphics_Vierecke;  Quadrat" rend="bitmap">148086</seg></emph></s> 
  <s type="es">A board game <del type="dnpc">exactly like</del><lb/> in fact chess but the<lb/> board has a
   square<lb/> which must never be<lb/> used&p.es;</s> 
  <s type="es">This may be</s><lb/> misleading&p.es;<lb rend="hl"/>
 <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,48v[5]" ana="date_19341000-19350500" rend="blbef_0 blaft_1" xml:lang="english">
  <s type="es" rend="indl_2">A board game in which only one man<lb/> <del type="dnpc">plays with</del> is said to play
   the other<lb/> to &usq.susq;answer&usq.eusq;&p.es;</s> 
 <emph rend="bl_1"/></ab>
 

<ab n="Ms-148,48v[6]" ana="date_19341000-19350500" rend="blbef_0 blaft_0" xml:lang="english">
  <s type="es" rend="indl_2">What if the other person always<lb/> correct<corr type="npcn">e</corr>ly described what I
   saw,<lb/> &amp.and; imagined, would I not say he knows<lb/> what I
   see&qm.eis;&dash;</s> 
  <s type="es">&udq.sudq;But what if he describes</s> 
 <pb facs="Ms-148_BC-r" rend="recto" n="pagename_Ms-148,BC-r pageref_Ms-148,99"/><fw type="pageno"/>
 <pb facs="Ms-148_BC" rend="recto" n="pagename_Ms-148,BC pageref_Ms-148,100"/>
 </ab>
    
     </body></text></TEI>
