Intellectualism Refuted (Part VII: Beyond Belief)

Gareth Evans (1982) argues that perception is belief-independent and in order to substantiate this claim he cites the well known example of the  Müller-Lyer optical illusion. He points out that despite any justified beliefs we may have about the actual lengths of the lines:  "[i]t will continue to appear to us as though, say, one line is longer than the other even when we are quite sure that it is not." Kelly (1998) endorses Evan's anti-intellectualist view that beliefs have no impact on our perceptions:

"Because perceptions are not subject to the canonical norms of rationality, then - because they are not, in other words, "rationally revisable" - they do not stand within the web of inferential relations that constitutes our beliefs, and ought not to be explained in terms of them."

For Kelly this web of inferential relations enables us to form beliefs, to make judgements and to infer conclusions. Despite the fact that there is no evidence to challenge this  conclusion, it remains a commonplace amongst philosophers to talk of "perceptual judgements" as if they are independent of "canonical norms of rationality." Evans is clear on this point. For him, judgement is connected with reasons and reasons are conceptual. If we are to avoid explaining perception by reference to "inferential relations" and "canonical norms of rationality" —of reasons— then something less extravagant than judgement is required.

We already have a candidate in Brook's theory of perception as a capacity to represent the things with which the  perceiver is engaged. In order to explore this conjecture it will be helpful to examine some frequently overlooked but nonetheless revealing anthropological research.

In one of the largest studies of its kind, Segal et al (1966) oversaw a global survey of cross cultural variations in susceptibility to optical illusions. The researchers found significant variance between differing communities and age groups. For example, some groups reported no difference between the lengths of the lines of the Müller-Lyer diagram. An earlier study by Hudson (1960), of culturally isolated South African children, encountered very similar findings. Both studies attributed the results to a lack of habitual exposure to pictures amongst the communities studied, and Hudson dubbed this: "pictorial illiteracy." In fact, even children well schooled in language and arithmetic skills (but lacking pictorial literacy) were not susceptible to what is commonly described as the "pictorial illusion of depth" and were therefore unsusceptible to the simulated spatial depth that many optical illusions exploit.

McCauley and Henrich (2006) write:

"For those who experience it, the illusion may persist, but susceptibility to the Müller-Lyer illusion is neither uniform nor universal. Moreover, a plausible argument can be made that through most of our species’ history most human beings were probably not susceptible to the illusion."

If this is true, and corroborating evidence can be provided from the art historical record as well as other sources (Deutscher 2010), then we have very good reason to suppose that our skills in the use of depictions are significantly implicated in the perplexities of such optical illusions. So, the conflicting responses we have when faced with the Müller-Lyer diagram are simply explained by the fact that we are disposed to represent the diagram in two different ways. We can treat the lines as a simulation of spatial depth or else we can treat them as two lines of equal length. No judgement need be imputed.

If the capacity to derive depth-cues from perspectival images is culturally acquired and is not an immediately available part of our genetically inherited perceptual repertoire then we have very good reason to suppose not only that perception is belief independent, as Evans claimed, but that perception is both non-conceptual and non-depictive. Language and depiction are skills that have developed through cultural evolution and both take time and practice to acquire. The capacity to imitate the behaviour of others on the other hand—to produce rudimentary versions of what Brook (1997) calls "Matching” representations—is an inherited skill upon which all of our more sophisticated communicative capacities supervene.

In Part VIII I will explore Kelly’s claim (pace Evans) that “Perceptual content is, sometimes at least, irreducibly articulated in terms of dispositions by the perceiver to act upon the object being perceived.” In doing so, I aim to explain how this gives strong support to the thesis that perception is fundamentally a communicative capacity: “a disposition to act” in representational terms “on [or with] the object being perceived.”

Intellectualism Refuted (Part VI: Do Animals Have Beliefs?)

"When we first begin to believe anything, what we believe is not a single proposition, it is a whole system of propositions. (Light dawns gradually over the whole.)" —Ludwig Wittgenstein

When Kelly (1998) examines Evans' suggestion that belief is a "sophisticated cognitive state: one that is connected with... the notion of judgement, and so, also, connected with the notion of reasons" he finds this "intuition" insufficiently threatening to would-be intellectualists. According to Kelly this allows intellectualists to maintain either that animal beliefs and perception are sophisticated conceptual states or else that animals have no beliefs because their perceptual states are non-conceptual whereas human perceptual states—or so the intellectualists might insist—are conceptual. Kelly makes a valid point, but I think Evans' intuition deserves closer consideration.

Any principled theory of perception must be consistent with an evolutionary account of the emergence of intelligent behaviours and as such it must be committed to a sparing view of the development of  conceptual capacities like beliefs. Evans was evidently wary of assuming that "sophisticated cognitive states" precede skills of concept formation. But perhaps what he most needed in order to substantiate this intuition was a theory of how conceptual skills might have evolved from other more commonly observable non-conceptual capacities. One possible component of such a theory can be discerned in the notion of demonstratives to which Evans also turns his attention. I will examine demonstratives later, with the intention of explaining how such utterances as "this," "that" and "here" gain their efficacy not as "demonstrative concepts", as some have argued (McDowell 1994, Loar 1997), but as conceptualisations of what are in fact non-conceptual skills of representation. For the moment though, it will suffice to say that skills of nonverbal representation (mimicry in particular) deserve to be considered as plausible precursors to sophisticated forms of symbolic communication and concept formation that would be impossible, or so I claim, without these prior capacities to produce, select and use non-conceptual (nonverbal) forms of representation.

Since the publication of Evans' book in 1982, the question of animal concepts has been the subject of widespread research. In 1994 Chater and Heyes undertook a meta analysis of the then extant research and found that: "[T]he idea of a concept has not been successfully decoupled from natural language, and hence there is currently no coherent account of what animal concepts might be." Since this time, no further evidence of animal concepts has come to light, and whilst it may be premature to suspend all enquiries, it appears that the assumption that animal capacities are intrinsically conceptual lacks substantive empirical grounds. Analytically speaking, the picture is little better. In 1982 — two years after Evans' death—Donald Davidson, whose work had a profound influence on Evans, published a widely cited essay, "Rational Animals," in which he argues that: "[I]n order to have a belief, it is necessary to have the concept of belief...[and]... in order to have the concept of belief one must have language." Like Wittgenstein (and perhaps Evans also), Davidson considered belief to be part of a system of propositions — the idea being that beliefs and rationality in general are acquired social traits.

It seems likely that Evans was motivated by similar considerations as those expressed by Davidson which suggests that his position was more substantive than Kelly gives him credit for. Nonetheless "Varieties of Reference" makes no mention of this reasoning and it is understandable therefore that Kelly finds the arguments unpersuasive. He is also right to point out that, in the face of a challenge to the supposed conceptual foundations of animal perception, intellectualists might simply insist that human perception is uniquely conceptual. Evans had an answer to this claim that I will explore next.

Intellectualism Refuted (Part V: Communicative Action Minus The Movement)

“A child who had never manifested in words, gestures, or play the working out of simple problems could not be said to work them out ‘in his mind’, any more than he could be said to know ‘in his mind’ the names of colours, if he was unable to say their names, or to point or to fetch the right colours when their names were called out. Thinking in ones mind (silent thinking, pausing to think) is not the most fundamental form of thinking, but instead presupposes thinking in play, work, or words.”

This passage from Norman Malcolm's 1977 essay "Thinking", emphasises the idea that mindedness is the consequence of skills acquired through active engagement with the world. Tempting as it is to conceive of the mind as a limitless inner workshop or skill-space for the ingenious generation of new capabilities, it is far better understood simply as a necessary component in all intelligent behaviour. But if this is so, then how is it possible for minded states to be executable independent of the actions of which they are necessarily part? As explained previously, intentional action is plausibly explained without reference to prior acts of predictive contemplation. In which case visualisation, anticipation, envisaging etc. are not prerequisites of intention or action. As Malcolm remarks: "Thinking in one's mind is not the most fundamental form of thinking." But, if we can act intentionally with no necessary antecedent acts of envisaging, then what is this more sophisticated form of thinking? How was it possible for mindedness to become stripped of its corresponding actions?

 "Thinking is movement confined to the brain." —Oleh Hornykiewicz

Working independently (in Australia and New Zealand), Donald Brook and Derek Melser have arrived at substantially the same conclusion on this question. Mindedness, they claim, develops as a result of learned competences in the social exchange of representations, i.e. communication. Such communication could include imitation, gestures, vocalisations, enactments and in our case: language. Mind is an inextricable feature of all communicative action, but some minded creatures—most notably humans—have become capable of inhibiting the movements that normally accompany action*. Mindedness then, is communicative action minus the movement.

This conclusion is well supported by evidence from developmental psychology. Early in the 20th Century, psychologists like Jean Piaget and Lev Vygotsky were the first to document commonly occurring patterns of language development in young children. They noted that preschool children pass through several phases of representationally directed action, beginning with imitation and then passing to "abbreviated demonstration," "speech-punctuated gesturing," "solitary speech" and finally "internalised speech." Something similar has also been observed in historical accounts of reading. According to Alberto Manguel (1996) the practice of "silent reading" was very uncommon until around the 10th Century. Prior to this time, reading was almost exclusively performed "out loud" and many wealthy readers even went so far as commissioning special reading rooms to which they could withdraw in order to read out their private correspondence etc.

These examples of internalised speech and silent reading underline the vital relation between the world involving skills of representation and their constitutive role in mindedness. The idea that it might be possible to read silently without first reading out loud or to plan one's actions "in one's mind" without first learning to intelligently negotiate obstacles, is not only a misunderstanding of the nature of skill and skill acquisition but a widespread misconception that continues to be a major source of confusion and philosophical argument.

In the next section I will explore the contention, attributed to Evans by Kelly (1998), that "perceptual content is belief-independent." Once again Evans' provides little in the way of evidence or argument, so the support for this view will have to be drawn from elsewhere. Nonetheless, there are  strong grounds for endorsing Evans' view that perception is belief independent.

*There is widespread evidence that many social animal infants engage in displays of mock aggression in which acts of biting etc. are only partially engaged. Such restrained displays presuppose the capacity for unrestrained aggression. Similarly, many higher apes engage in abbreviated gestures that are sufficient to elicit the same responses from conspecifics as would be expected from unabbreviated gestures.

Intellectualism Refuted (Part IV: Unanticipated Intention)

"Theories of mental content attempt to solve what is often referred to as Brentano's problem: the problem of explaining intentionality, explaining how mental states can be about things or be directed on to things in this way." —Karen Neander 2012

Previously I discussed the problems that arise as a consequence of conceiving of brains as representation containers as opposed to evolved organs of embodied response. I also raised some related objections to the notion of aboutness that many theorists claim is fundamental to the explanation of intentionality. My contention is that representations can only possess aboutness by virtue of their participation within an already intentionally directed system of rules. Without a rule to govern its use, even the most basic symbol is incapable of standing-in for anything (Bickhard and Turveen 1995, Bickhard 2004). If this is so, then the problem of explaining intentionality, raised by Neander in the above quote, runs into severe difficulties. Furthermore, if an inner representation is required to initiate an intentional action then this representation must itself be either intentional or unintentional. If it is intentional then it is a necessary condition that it also be initiated by another intentional representation and so on ad infinitum. If, on the other hand, the initiating representation is unintentional then the resultant behaviour cannot be intentional. I conclude therefore that inner representations are logically incapable of underwriting any form of intentional action — human or animal.

So, how can we answer "Brentano's problem" whilst avoiding the perils of infinite regress? In his “Philosophical Investigations,” (1953) Wittgenstein writes: "When people talk about the possibility of foreknowledge of the future they always forget the fact of the prediction of one's own voluntary movements.” Wittgenstein was concerned here with the question of why we are not surprised by our own voluntary movements. Despite the fact that we need not envisage any movement of our body prior to actually moving, we are invariably unsurprised by the outcome. As Wittgenstein suggested, and Gilbert Ryle (1949) further explored, where such voluntary actions are concerned, the action and its anticipation are one and the same thing. In other words, enacting a learned action is in itself predictive of the action's unfolding and therefore no prior act of contemplation is necessary for the behavior to qualify as being intentional. The learned capacity to do something presupposes its outcome. Another way of putting this would be to conceive of kinaesthetic learning (the learning of movements) as the acquisition of dispositions to move in certain ways in certain circumstances. These acquired dispositions are then capable of underwriting future intentional behaviour of the same kind.

If the above is correct, then there may be grounds to suppose that animals are capable of intentional action. Any questions we might have concerning anticipation and mindedness will have to wait until the next post, but what of perception? Was Gareth Evans correct in his intuition that humans and animals share certain perceptual characteristics? Gilbert Ryle (1949) argued that perception is less an action or procedure than an "achievement," and as Derek Melser (2009) remarks: "achievements necessarily imply prior active strivings." Melser contends that perception is a skill that depends on language acquisition and is developed through interaction with other perceivers in a process that he calls "tokening and concerting". On this view, perceivers engage in active dialogue about perceived objects that gradually accumulates into a perceptual repertoire of associated linguistic and behavioural responses. Melser also offers an important refutation of the commonplace assumption that perception is a relation between acts of perceiving and things perceived. He writes: "The question of what thing X is has the same answer as the question of what perceiving X is. [...] Hence they are not two things and cannot be 'related.'"

If language is essential to the development of perceptual skills as Melser claims, then Gareth Evans' view that animals and humans share certain perceptual states would seem to be in difficulty. Nonetheless there may yet be hope for anti-intellectualism. Like Melser, Donald Brook (1969, 1997, 2014) emphasises the cultural basis of perceptual skills but unlike Melser, he claims that perception is not exclusive to language-users but may be reasonably attributed to other creatures capable of producing nonverbal representations of the things with which they are engaged. The challenge for this theory is to provide evidence of these hypothesised causally influential but untriggered dispositions to represent. However, if it is true that many creatures are also capable of learning new behaviours, as is widely observed, then it follows—at least in principle—that they are capable of performing these skills and demonstrating them to other perceivers in just the way that Brook suggests. Indeed, if an animal learns a behaviour by observing and imitating the actions of a parent or sibling, then we have all the evidence we need that such an animal is a perceiver by virtue of its manifest capacity to imitate (to represent) this observed behaviour.

So the non-conceptual view that Gareth Evans is often invoked to defend would seem to have healthy grounds for support — so long as we bear in mind the importance of nonverbal know-how. As the arguments develop, we will see that this Rylean approach presents yet more significant challenges for intellectualism.