From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

1. Conceptual Metaphors

Decades after Wittgenstein’s death, work in diachronic cognitive linguistics identified a major process of language development: metaphorical extension (Traugott & Dasher 2005). We tend to conceptualise unfamiliar or abstract matters in terms initially applied literally to concrete or familiar things or actions (‘grasp a stone to throw’), whose use we extend metaphorically for application in abstract or new contexts (‘grasp the implications of a claim’). Such extension is the single most important process whereby languages become equipped to deal with the abstract. For example, much of the English mental vocabulary is recruited in this way from the domains of manipulation and perception (Jäkel 1995).

Typically, metaphorical extension is wholesale: The use of a whole set of related terms is extended from the initial (‘source’) domain to a new (‘target’) domain. Thus, terms initially applied in talk about visual search came to be employed, wholesale, in talk about goal-directed intellectual efforts: efforts to solve problems, answer questions, explain facts, events or actions, etc. Thus we say about efforts to understand actions:

• It is clear or obscure to me why you did what you did, according to whether or not I manage to see any reasons for acting that way. I may look for reasons where these are hidden or be blind to reasons that are in plain view. An illuminating explanation which throws new light on your action will let me see reasons I had previously overlooked, and thus get a fuller picture of these reasons, or at least let me catch some glimpse of them, where I was previously completely in the dark. A fresh look at the situation to which you responded may reveal threats in whose light your action no longer looks as out of character as it did at first sight.

Such wholesale extension of terms preserves inferential relations between the several terms involved. Whether we are talking about swallows on the roof or reasons to act, you can only ‘point out’ to me what you ‘see’ yourself, so that ‘S₁ points y out to S₂’ entails ‘S₁ sees y’, while for both birds and reasons ‘x is hidden’ entails that neither you nor I can see x, etc. etc. (see Fischer 2010, ch.4 for details). The result is a

• conceptual metaphor: a systematic mapping of terms from a source- to a target-domain, which preserves relations between them.

Whether applied to elements of the source domain of visual search or elements of the target domain of intellectual effort, the perception-related terms at issue stand in the same inferential relations to each other. This preservation of inferential relations forges a structural analogy between visual search and intellectual effort:

• A is structurally analogous to B iff a set of elements of A can be mapped onto a set a elements of B, which stand in some of the same inferential or other relations to each other.

Various related processes of metaphorical extension forge a series of structural analogies between perception and intellectual activities and achievements including reflection and knowledge. More generally, such extension systematically forges structural analogies between the target- and the source-domains of conceptual metaphors, between the more concrete and the more abstract.

These structural analogies are not ‘false’; they do actually obtain. But they can be said to have been ‘accepted into language’. And we shall presently see that they are ‘at work in the unconscious’, namely in non-intentional analogical reasoning in which even the most competent thinkers unwittingly but systematically presuppose also further, material, analogies, which are ‘false’.

2. Non-intentional Analogical Reasoning

Structural analogies play a crucial role in non-intentional analogical reasoning: When thinking about one thing, we may unwittingly seize on a structurally analogous thing as a model and spontaneously project properties of the model onto the other thing (the target), without being aware of using anything as a model or making any analogical inference. Under certain circumstances, thinkers are prone to make analogical inferences they are not aware of making.

• A thinker makes a non-intentional analogical inference iff he spontaneously makes an inference that presupposes that some thing (the target) is in some respect like a structurally analogous other thing (the model), while unaware of presupposing this or invoking any model.

This happens, in particular, when

• (1) thinkers pursue no practical goal,
• (2) thinkers lack knowledge of relevant detail or context, and
• (3) a structurally similar familiar model is available.

Much philosophical reflection satisfies these three conditions. First, philosophical reflection typically not directed towards any practical goal. Second, it often involves swift generalisation or general reasoning without detailed reference to specific examples and is couched in technical terms defined without any such reference. This leaves thinkers adrift without much knowledge of detail or context, including details and features of pertinent contexts that would be relevant for determining the truth or falsity of philosophical claims they consider. Third, much philosophical reflection is about abstract matters for which pertinent models are made available by the process of metaphorical extension we have considered a moment ago: The systematic extension of terms from more to less tangible and public matters systematically forges structural analogies between simple and familiar domains (like the domain of visual search) and more abstract domains (like the domain of intellectual effort). Thus one major process of language development generates a rich store of simple models of the abstract, ready to be unwittingly seized on the moment we engage in unduly general reasoning about abstract matters and are not guided by any specific goal (Fischer 2008b).

In the experiments that established the existence of non-intentional analogical reasoning, subjects assimilated targets to models more extensively than they had warrant to do (Day & Gentner 2007). This may happen even against better knowledge: That we spontaneously make an inference which presupposes that the target is in some respect like the model, while unaware of presupposing this, implies that we may presuppose that the two are alike in ways in which we do not want to assimilate them, including ways in which we know the two to be different. Such spontaneous inferences strike those who make them as intuitively compelling (Fischer 2008a: 55-9). Thus we may come to find compelling a conclusion that presupposes that two things are alike in ways in which we know them to be different. Unwittingly and against better knowledge, we may excessively assimilate one thing to another, presuppose material analogies between a target and a merely structurally analogous model, e.g., between merely structurally analogous elements of the target- and the source-domains of conceptual metaphors.

All of us know full well, for example, that ‘I see a swallow’ implies a bird is around to be seen, that its location is within my field of vision, i.e., within suitable range of my eyes, and that I see it with my eyes. By contrast, ‘I see your reasons’ or ‘I look at the issue from all sides’ obviously do not imply anything of the sort: Neither the issue nor your reasons need be around to be looked at or seen, within appropriate range, with an organ of sense. But, under the circumstances considered, thinkers may go along with leaps of thought presupposing, e.g.

• (R) To think about, ‘consider’ or ‘look at’ something is to perceive some thing somewhere, somehow.

We have seen that this phenomenon, the excessive application of conceptual metaphors against better knowledge, is captured quite well by Wittgenstein’s characterisation of philosophically pernicious pictures and analogies implicit in language (cp. end section 1). Let’s define:

• S is in the grip of a philosophical picture iff in non-intentional analogical inferences S unwittingly presupposes material analogies between source- and target-domains of a conceptual metaphor.

3. Positing ‘Minds’

Non-intentional inferences presupposing (R) and further material analogies to the model of visual perception led thinkers to posit in us a space and organ of inner perception, called ‘the mind’. In the grip of various perceptual pictures, early modern philosophers replaced ‘rational’ and ‘sensitive souls’ by ‘minds’ (a concept with a new extension and intension, cp. Kenny 1993).

The idea that there are such spaces of inner perception is rendered compelling by leaps of thought which proceed from truisms about the conventional metaphorical use of perception-verbs in which they are ordinarily used in talk about intellectual activities and achievements. When we say, for example, that someone ‘looks at’ an issue or ‘contemplates’ whatever he thinks about, we are precisely not speaking about visual perception of physical objects in our environment:

• (T₁) To ‘consider’, ‘look at’, think about something is not to perceive any thing somewhere around us, with our eyes (or any other of our five senses).

But many philosophers found it intuitively compelling to go along with an inference which presupposes (R) above, and add:

• (C) To think about something is not to perceive anything around us with our external sense-organs; it is to perceive things within us, with a further, inner, sense.

Analogous conclusions about knowing, understanding, or remembering may strike us as compelling, even though they are patently at odds with obvious facts or acknowledged definitions. As a result, thinkers spontaneously make various moves to accommodate the claims in view of such conflicts, and explicitly maintain only the results of these spontaneous moves. The present conclusion, for instance, is at odds with the conceptual truism:

• (T₂) To ‘consider’, ‘look at’ or think about something is not to perceive anything in our bodies, and is not to employ any bodily organ of sense.

The most common initial response to such conflicts is spontaneous reinterpretation of the problematic conclusions. Thinkers ‘sublimate’ what they want, but find inappropriate to, maintain, ideas they find compelling but which just won’t do.

• A thinker sublimates a claim iff he spontaneously places a new interpretation on its key terms, so as to be able to maintain its expression in the face of a conflict with claims he accepts.

Frequently, such sublimation is metaphorical; we are told to interpret talk of things being ‘perceived’ and ‘existing’ within us figuratively rather than literally: ‘[W]hen I speak of objects as existing in the mind … I would not be understood in the gross literal sense, as when bodies are said to exist in a place’ (Berkeley 1734: 250; cp. Locke 1700: II.ix.10). The gist of many such explanations can be summed up thus:

• (E) To ‘consider’ or think about things is to ‘perceive’ something in a metaphorical sense, in an ‘inner space’ that is not literally, physically in us, with an ‘inner sense’ not to be confused with any bodily sense-organ.

Until today, the label “the mind” is frequently used to refer to that other than physical ‘inner space’. Locke’s contemporaries simultaneously used it to refer also to the other than bodily ‘organ of sense’ with which we ‘look at’ and ‘see’ the things ‘in’ that ‘inner space’:

• (D₁) “the mind” = def.: (a) the inner space without physical extension or location in which we perceive the things we do not perceive with our five senses; (b) the non-bodily inner organ of sense with which we perceive the objects in that space.
• (D₂) “idea” = def.: ‘whatsoever the mind perceives in itself’, i.e. (resolving the metonymy), whatever is perceived in the mind, with the mind (Locke 1700: II.viii.8)

Thus “the mind” is introduced as a label for the perceptual space and organ we posit when we find conclusions like (C) compelling, want to maintain them in the face of their conflict with truisms like (T₂), and therefore immediately resort to maintaining them in a metaphorical sense, through explanations like (E). The overly literal application of perceptual metaphors has us posit a perceptual space and organ in us; metaphorical sublimation removes the two from our bodies and turns them into ‘minds’.

This is the first move in a long struggle with the tendency to excessively assimilate a wide range of intellectual and other target-domains to the source-domain of perception. This struggle manifests itself in lines of thought which, on the one hand, rely on tacit assumptions that treat the mind as a literally inner and physical space of perception and, on the other hand, simultaneously invoke explicit assumptions to the contrary, viz. sublimating explanations of different kinds. Such lines of thought then lead to the conclusions that the mind is a non-physical, private and transparent realm (Fischer 2009, ch.5). The apparent clash of these conclusions with a scientific world-view (and with truisms about various intellectual activities and achievements) gave rise to classical mind-body problems.

4. Wittgenstein Vindicated

‘It is pictures rather than propositions, metaphors rather than statements, which determine most of our philosophical convictions’ (Rorty 1980: 12). Proponents of this view typically focus on metaphors deliberately employed and on pictures actually endorsed by the philosophers guided by them, like the picture of the mind as a space of inner perception, or repository of pictures (which Locke explicitly endorses e.g. in II.xi.17) – or (more recently) as a telegraph exchange or computer. This paper developed Wittgenstein’s suggestion that philosophical reflection may be shaped, more fundamentally, by ‘similes at work in the unconscious:’ by (conceptual) metaphors we employ without being aware of doing so. In non-intentional analogical reasoning we apply such metaphors more literally than we know appropriate, presuppose material analogies we know not to obtain, and are thus led to conclusions which engender philosophical perplexity. These findings vindicate Wittgenstein’s further suggestion that such perplexities can be resolved by tracing them back to false analogies as their source: We can, for example, resolve perplexities engendered by the conception of the mind as a private space of perception by showing its proponents that their conception relies on ‘false analogies’ like (R), which they explicitly reject in giving what we called ‘sublimating explanations’ (like E). By reconstructing the non-intentional reasoning that decisively shapes and pre-structures philosophical reflection, we can unearth inferences that are unsound not only by the lights of some philosophical critic but of the very philosopher who unwittingly makes that move – against better knowledge.