Is Non-Conceptual Content a Kind of Given?

We can go back to the 18th century and to Thomas Reid to see what appears to be a reference to what is now called “non-conceptual content”. Reid wrote:

“[S]ensation, taken by itself, implies neither the conception nor the belief of any external object. [] Perception [on the other hand] implies an immediate conviction and belief of something external.”

Now let's jump to the first half of the 20^th century, when “sense-data theory” was in fashion. An experience of a sense-datum was seen in non-conceptual terms. It is of course true that this position was problematic in many ways. As Michael Williams puts it:

“It is important to see that acquaintance with sense-data is 'direct' or 'immediate' in two senses. Not only is it independent of any further beliefs, it is pre-conceptual. It makes propositional knowledge (involving conceptualisation) possible.”

Now we can jump forward to another distinction offered by Laurence BonJour. He argues that what he calls “sensory content” is not in fact “propositional or conceptual in character”.

*************************************

If we return to the sense-datum of a red patch.

That must have meant (i.e., to sense-data theorists) that a red patch sense-datum is firstly experienced and only then perceived as a red patch. That is, the concepts [red] and [patch] are applied to the red patch. (What is meant by “applied” here?)

Thus, the sense-datum of a red patch is a basis of later knowledge, not an example of knowledge itself. It can become conceptual. So there's no reason to believe it is incorrigible, direct, immediate, pure or anything else like that. The Given is only seen as the Given after the fact.

Following on from that, this raises the possibility (or actuality) that even if we accept non-conceptual content, we may still go wrong when we describe it. However, how would we (or anyone else) ever know that? Here we have another problem: the problem of private mental states. (In this case, private sensory content.)

As for BonJour's position.

We can say that such “sensory content” can become conceptual content. And only then can it (as Michael Williams puts it) “justify basic beliefs”. More simply: sensations cause beliefs (i.e., they aren't themselves beliefs). This is a position which will be defended later.

This question must now be asked:

What is the precise nature of the movement from non-conceptual content to conceptual content?

Animals

It can be argued that some animals have the same “non-conceptual content” as human persons – at least in certain cases. It's just that we can (or do) apply words or concepts to that very same content. This is a position expressed (if not endorsed) by Alex Byrne when he states the following:

“Some of the perceptual states of lower animals have contents in common with human perceptual states.”

We can of course ask (in tune with philosophers like Daniel Dennett) how we could ever know (or even surmise) that “animals have contents in common with human perceptual states”. On the basis of their behaviour? (How would that work?) On the basis of their evolutionary proximity to human persons?

In any case, Martin Davies also allows animals non-conceptual content when he says that “human infants and certain other creatures” are “arguably [ ] not deployers of concepts at all”. More specifically, this perceptual content is free “of any judgement that might be made”. 

We can also take the position of Brian Loar.

Let's talk about a dog called Fred.

According to Loar (not his own example), Fred “picks out a kind” - the kind human beings call “human beings”. More correctly, Fred picks out a particular (say, Mike) and sees that he/it belongs to the kind we call (though it doesn’t) human beings. At no stage of the game is Fred’s x is F anything like our x is F. That is, it's not linguistic or sentential.

So how close must Fred’s x is F to our x is F? More relevantly, why do we demand an exact parallel with our x is F in order to allow attributions of concepts and beliefs to dogs and other animals?

Having put the dog's position, it doesn't appear to be an argument for non-conceptual content in that Loar says that Fred (the dog) “picks out a kind”, etc. Clearly, noting or picking out kinds (even if non-linguistically, etc.) doesn't seem like a candidate for non-conceptual content.

Nonetheless, something (or some things) must also come before the linguistic expressions of human persons: both as a species and as individuals. Our linguistic expressions didn’t occur ex nihilo.

Here’s Paul Churchland making related points:

“[L]anguage use is something that is learned, by a brain already capable of vigorous cognitive activity; language use is acquired as only one among a great variety of learned manipulative skills; and it is mastered by a brain that evolution has shaped for a great many functions, language using being only the very latest and perhaps the least of them. Against the background of these facts, language use appears as an extremely peripheral activity, as a species-specific mode of social interaction which is mastered thanks to the versatility and power of a more basic mode of activity. Why accept, then, a theory of cognitive activity that models its elements on the elements of human language?”

Peacocke, Davies and Tye

One direct case of purportedly non-conceptual content is offered up by Christopher Peacocke when he writes the following:

“Only those with the concept of a sphere can have an experience as of a sphere in front of them."

In detail:

“The natural solution to this... quandary is to acknowledge that there is such a thing as having an experience of something as being pyramid shaped that does not involve already having the concept of being pyramid shaped.”

This is problematic in that there may still be concepts involved in this experience. That is, it may still be what Peacocke calls “object-involving”. In any case, this person (or animal) has an experience - and even an experience of a something (x). It's just that he (or it) doesn't have an experience “as of a sphere”. It's a sphere to us; though not to him (or it).

Elsewhere, Peacocke says that the

“content of experience is to be distinguished from the content of a judgement caused by the experience”.

In our example, this would be a judgement that x is a pyramid.

In basic terms, the “content of experience” and the“judgement” don't occur at precisely the same time. Whether this temporal way of looking at things makes sense (or is acceptable) is another matter. I say that because it can argued that the experience and judgement occur at one and the same time.

Peacocke also sets up a relation between the non-conceptual and the conceptual when he says that

“thought can scrutinise and evaluate the relations between non-conceptual and conceptual contents and obtain a comprehensive view of both”.

This question must again be asked here:

What is the precise nature of the movement from non-conceptual content to conceptual content?

Colin McGinn points out the position represented by philosophers like Peacocke when it comes to “representational content”. McGinn says that they accept

“prerepresentational yet intrinsic level of description of experiences: that is, a level of description that is phenomenal yet noncontentful".

Peacocke himself says that “sensational properties do not determine representational content”. (Peacocke cites an example of an array of dots which can be seen as either vertical or horizontal rows^{1 and} Wittgenstein's duck-rabbit seems to be relevant too.) Concepts are part of this story. 

More specifically on representation, Peacocke says that the

“representational content of a perceptual experience has to be given by a proposition, or set of propositions, which specifies the way the experience represents the world to be”.

Does Peacocke mean something linguistic or something abstract here? Does he mean that these propositions need to be articulated or verbalised? And why do we need propositions at all in these cases?

Peacocke also appears to make a mistake about a supposedly non-conceptual experience.

He says that a person “waking up in an unfamiliar position or place [will have] minimal representational content” . Yet surely unfamiliarity doesn’t entail lack of conceptual or “representational content”. This person may still conceptualise his “unfamiliar position or place”. Perhaps this only displays Peacocke’s linguistic or propositional bias. This person may not describe or form “propositional judgements” about his unfamiliar position or the new place he finds himself in. However, that may be irrelevant to conceptual and/or representational content.

To summerise Peacocke's position:

sensation ⟶ perception ⟶ judgement

Martin Davies too makes an explicit distinction between what he calls the “perceptual content of experience” and the having (or possession) of concepts. In detail:

“[T]he preconceptual content of an experience is a kind of non-conceptual content. What this means is that a subject can have an experience with a certain perceptual content without possessing the concepts that would be used in specifying the content of that experience.”

Thus, in Davies's jargon, the “perceptual content of experience” comes first, and only then are concepts applied to it. Indeed, it seems possible that this perceptual content of experience may remain (as it were) free of concepts.

Davies's perceptual-content idea is even more radical in that it “is not object-involving”. The basic point here seems to be that non-conceptual content can't (by definition?) be object-involving.

Martin Davies also divorces perceptual content from the “representational”. That is, the non-conceptual implies (to me at least) the non-representational. According to Davies, that non-conceptual “substrate” is added to by a “representational superstructure”. In addition, the “sensational” is also distinguished from the “representational”.

Finally, Michael Tye also makes a temporal division when he says that

“visual sensations feed into the conceptual system, without themselves being a part of that system”.

One can now ask how “visual sensations” can “feed into the conceptual system” without having some kind of vital connection to that system. And if they have such a connection, then can they still seen as non-conceptual? That is (as asked twice before), what is the precise connection (or link) between the non-conceptual (visual sensations in this case) and the conceptual system? One is tempted to infer (as with John McDowell) that in order for these visual sensations to be able to feed into the conceptual system, then they must share something with that system. Otherwise how does that system distinguish irrelevant sensations from relevant ones?

Despite that, according to Tye, “phenomenal content” is by definition non-conceptual. And he too (like Peacocke) gives us his own division between “phenomenal content” and belief. Thus we have:

“A content is classified as phenomenal only if it is nonconceptual and poised [for use by the cognitive centres].”

And:

“Beliefs…[which] lie within the conceptual arena, rather than providing inputs to it.”

Here “phenomenal content” is seen as input to be worked upon later. Beliefs, on the other hand, are outputs. Is this phenomenal content “the Given”? It certainly seems to be when Tye made the epistemic point that phenomenal content is “poised for use by the cognitive centres”. That is, it comes (epistemically) before such “use”.

We also have an explicit tying of concepts to language from Tye when he writes the following:

“Having the concept F requires, on some accounts, having the ability to use the linguistic term 'F' correctly.”

Yet Tye also cites mental content that is non-linguistic when he says that

“after-images, like other perceptual sensations, are not themselves thoughts or beliefs; and they certainly do not demand a public language”.

It's very hard to see afterimages as conceptual. Then again, afterimages are a very special case of mental content. That is, afterimages don't seem to be relevant to this discussion because they're unlikely (or rarely) to be the basis of later conceptual content.

Critics of Non-Conceptual Content/the Given

Wilfrid Sellars

Wilfrid Sellars (in his 'Epistemic Principles') lays his critical cards on the table when he said the Given

“would be a level of cognition unmediated by concepts; indeed it would be the very source of concepts”.

Sellars was right to imply here that there can't be a “level of cognition” which is “unmediated by concepts”. However, we needn't also conclude (or accept) that these sensations can't be a “source of concepts”. The problem is that Sellars fuses cognition with non-conceptual content. Yet the two needn't go together. Indeed not even an old-fashioned believer in the Given would have believed that. And isn't that why non-conceptual content is a (to use Sellars' own word) “source” of cognition and concepts, not an example of these things?

Incidentally, Sellars himself did make a distinction between the two when he said that a “sensory element [of perceptual experience] is in no way a form of thinking”. So Sellars happily concedes a “sensory element”. Having said that, that sensory element may not come first. It may simply be part of the perceptual experience from the very beginning.

John McDowell

John McDowell explicitly believed that the acceptance of non-conceptual content is effectively a rebirth of “the myth of the given”. McDowell is now well-known (i.e., within epistemology) for holding this position. That is, for his rejection of the idea that we have a temporal division (as I see it) between the mind being presented with a non-conceptual Given, and then a later application of concepts to that Given.

McDowell himself writes (in his Mind and World):

“[T]he content of a perceptual experience is already conceptual. A judgement of experience does not introduce a new kind of content, but simply endorses the conceptual content, or some of it, that is already possessed by the experience on which it is grounded.”

What does McDowell mean by “simply endorse the conceptual content” (i.e., apart from claiming that the content is already conceptual)? What does McDowell mean by “already conceptual” here? How does a concept-based “judgement” differ from the already conceptual “content of a perceptual experience”? And isn't all this dependent on what philosophers take concepts to be?

So McDowell believes that we're fooled into believing that there's such a thing as non-conceptual content. In the case of “rich” experiences (such as when we don't have the words or concepts for particular shades of colour), it's nonetheless the case that we have a “a recognitional capacity, possibly quite short-lived, that sets in with the experience”. (This is similar to the stress which David Lewis made on “recognitional capacites” when discussing the what-Mary-didn't-know scenario. That is, Mary doesn't "acquire any new facts". However, she does acquire new recognitional abilities.)

Despite all that, it's hard to see a situation in which having a “recognitional capacity” can ever be mistaken for a non-conceptual state by so-called “non-conceptualists”. After all, McDowell argues that it's fully conceptual. So perhaps McDowell is wrong to assume that this is an example of a conceptual state mistaken (by non-conceptualists) for a non-conceptual state. In other words, don't non-conceptualists have something more basic in mind when they refer to non-conceptual content?

A Rich Experience

It's often said that an experience which is “rich” can't be entirely conceptual. For example, Alex Byrne quotes a fictional person saying:

“It appears to me that my environment is thus-and-so.”

And who then says:

“So I suppose the content of my experience is rich/perspectival/phenomenal/non-conceptual.”

I'm not sure if this is a good way of putting it because if the environment is seen to be "thus-and-so”, then doesn't that imply that it's not “non-conceptual”?

Anyway. The above is a simple way of saying that this person had an experience of a particular environment which he couldn't completely describe. Or, rather, at the actual time of the experience he didn't describe it; though afterwards he may well have been able to do so. Nonetheless, even after the experience, there would still have been elements of that environment for which he has no words or concepts.

So does this show that this experience was at least partly non-conceptual?

Here's another description of a rich experience from Christopher Peacocke:

“Our perceptual experience is always of a more determinate character than our observational concepts which we might use in characterising it. A normal person does not, and possibly could not, have observational concepts of every possible shade of colour...”

Can we say that an experience is “more determinate” even if it's not conceptualised in any way? In what sense, then, is it determinate? Here there's a hint at a kind of discrimination which doesn't involves concepts, let alone words or descriptions.

Despite that, even if a person has no “observational concepts” or words for “every shade of colour”, that person is said to still note the unnamed shades of colour. But how is that possible without concepts? Perhaps the problem is tying concepts too closely to public words. Surely an animal (say, a dog) can discriminate without public words or “observational concepts”.

Bill Brewer articulates this point in the following:

“For surely a person can discriminate more shades of red in visual perception, say, than he has concepts of such shades, like 'scarlet', for example.”

That seems to be the case. However, according to Brewer, the “conceptualist” has an answer to this. It is to

“exploit the availability of demonstrative concepts of color shades, like 'that_r shade', said or thought while attending to a particular sample, R”.

As can be seen, this is still language-fixated in that although there are no public words for these colour shades, this person is still saying “that_r shade” to either himself or to another person. Surely this rules out any discriminations an animal may make.

So here we may have a “fineness of grain” (or “richness”) without concepts or judgements that implies a level of discrimination which occurs without concepts - or at least without words.

Of course one can apply concepts or words after the fact. One can even invent one's own neologisms for experiences or colours one doesn't know the name of. But we may still have had an experience of the colours without using concepts or words.

We can now go beyond talk of the different shades of colour and say, as Gareth Evans did, that perception itself always (or often) has a “phenomenological richness” which goes beyond the concepts used in perception. In other words, experiences or perceptions are more fine-grained than can be accounted for by simple references to the many different shades of colour. Indeed, phenomenologically, an experience is almost infinitely rich (or detailed). And even if we had public words for everything within it, such words would still never be used during the actual experience itself. 

Davidson on Causes and Sensations

It may be useful to press-gang Donald Davidson into this debate.

Whereas we can stress non-conceptual content, Donald Davidson himself stressed the “causes” of what he called “sensations”. So, in this picture, the causes of sensations can be said to take the place of non-conceptual content.

In the following passage (from his paper 'A Coherence Theory of Truth and Knowledge') Davidson wrote:

“The relation between a sensation and a belief cannot be logical. Since sensations are not beliefs or other propositional attitudes. What then is the relation? The answer is, I think, obvious: the relation is causal. Sensations cause some beliefs and in this sense are the basis or ground of those beliefs. But a causal explanation of belief does not show how or why the belief is justified.”

So it's worthwhile rewriting this passage for clarification and to put it within the context of non-conceptual content. Thus:

The relation between “sensations” (or non-conceptual content) and beliefs cannot be logical. Since sensations are not beliefs. What then is the relation? The answer is, I think, obvious: the relation is causal. The world causes non-conceptual content (or sensations) and in this sense is the basis or ground of concepts (or conceptual content) and beliefs. But a causal explanation of non-conceptual content (or sensation) doesn't show how or why judgements about it are true or false of the world.

Another passage (from the same paper by Davidson) is even more apposite in this context. Davidson wrote:

“Accordingly, I suggest that we give up the idea that meaning or knowledge is grounded on something that counts as an ultimate source of evidence. No doubt meaning and knowledge depend on experience, and experience ultimately on sensation. But this is the 'depend' of causality, not of evidence or justification.”

Here again we can rewrite Davidson in the context of our take on conceptual and non-conceptual content. Thus:

I suggest that we give up the idea that conceptual content (or belief) is grounded on something that counts as an ultimate reality: either non-conceptual content or the world. No doubt conceptual content (or beliefs) ultimately depend on non-non-conceptual content (sensations) or the world. But this is the 'depend' of causality, not mind-independent truth or fact.

In Davidson's terms, conceptual content causally “depends” on non-conceptual content (or sensations) and the world (or causes). However, that non-conceptual content or world doesn't - and can't - in and of itself guarantee us truth. Thus it can't be seen as the Given in the 20th century sense.

In that case, perhaps we can say that our causal interactions with x are the Given; though not the beliefs these causes bring about. In that case, as Susan Haak says, it is

“only propositions, not events [or objects], that can stand in logical relations to other propositions [or beliefs]”.

The causal interactions (or non-conceptual content/sensations) themselves are neither beliefs nor propositions. Therefore they can't “stand in logical relations to other propositions” or beliefs. The causal interactions (or sensations) are causes of beliefs; though, in and of itself, they are neither evidence for such beliefs nor justifications for further beliefs.

In any case, the same causal context - taken only in itself - can cause different beliefs in different people and possibly different beliefs in the same person at different times. The interpretations of our causal contacts depend on our prior beliefs and the prior concepts which we apply to our causal interactions. And even if a particular causal contact brings about the formulation of new beliefs or new concepts, these will still be dependent upon - or be related to - prior beliefs and prior concepts.

Conclusion

To sum up. It can be said that surely there must be some kind of Given in order to get the ball rolling.

Thus in Davidson's scheme we had:

causes ⟶ sensations ⟶ beliefs 

instead of the more basic:

sensations (or non-conceptual content) ⟶ beliefs (or conceptual content)

Alternatively:

i) An experience of x.

ii) Then an experience of a [?].

Or:

i) Sense experience x.

ii) Then sense experience x + conceptual content (or plain concepts).

We can now say that i) and ii) may not, or cannot, occur at one and the same time.

Thus “the Given” needn't remain given. That is, i) becomes ii).

************************************

Note:

1) 

Christopher Peacocke's example of a "sensational substratum" of either three vertical columns or four horizontal rows doesn't seem to work. He says that "we see the array as three columns of dots rather than as four rows". However, isn't that simply because the dots in the columns are closer together than the dots in the rows? If the distances between the dots were identical in both cases, then what would we "see"? The image at the top of the page is more balanced than Peacocke's own example in this note.