I Am Therefore I Think

21 01 2007

There is a rather entrenched notion within the world of psychological philosophy that in order to truly understand an idea we must gain a thorough understanding of the semantics of that idea’s expression. Put in other words, an awareness of language must necessarily precede an awareness of the ideas put into words. Can ideas not put into words be understood? Better still, are they even ideas? This question is a powerful one and, in many respects, it lay at the heart of Orwell’s frightening classic, 1984. If our languages are hijacked and words like “freedom” robbed of their true meaning (with concepts like “freedom” being thus bereft of linguistic expression), the ideas that were previously communicated with these words will simply cease to be. Of course, this is impossible; but not for the obvious reason.

The obvious reason might be that such concepts can exist without the arbitrary label to which they had been joined, but this is not necessarily correct. I would conjecture that a person living in utter isolation throughout their entire lives would still develop labels for things (whether articulate or inarticulate) and that it is those labels that grant them the ability to think about those particular objects. Without the labels, their thoughts would always be ‘simple’ – by which I mean to imply that they would not develop beyond that which requires no prior knowledge. It is only by using ‘names’ for things that we are able to think about those things themselves; that we are able to actually encapsulate them and appreciate them from without. By destroying the word “freedom”, we destroy the concept “freedom”.

Nonetheless, Orwell’s nightmarish vision was unrealistic in respect of the fact that labels for outlawed concepts would simply be reinvented from scratch over time. Labels are continually being invented and, in many instances, it is the label that precedes the object to which is being referred. Mathematical logic is a primary example, within which labels might be deduced on the basis of calculations and only later may their referent be fully understood in its own right. In common experience, however, labels tend to be invented as the necessary adjunct to the delineation of a new idea, but are nonetheless developed at a rapid pace. Just as new ideas are continually being brought into existence, so too are new terms continually being either reapplied or invented from scratch.

One might work backwards from this premise, as many scholars of psychological philosophy do, and argue that the idea itself can not be approached except through a thorough semantic appreciation of the means by which that idea is conveyed. This assertion is not merely a pragmatic one, but actually strikes at the heart of cognition itself. Not only must we be forced to understand the means by which something is explained to us, but that thing itself cannot be understood even by those who have already grasped it except through the medium of linguistic expression. This is known as the Thought-Language Principle.

This principle is not entirely undebated. José Luis Bermúdez, Professor of Philosophy and the Head of the Department of Philosophy at Stirling University, writes about one reason as to why this principle may need to be re-evaluated. His book, The Paradox of Self-Consciousness, explores a particular conundrum that only exists if we posit the Thought-Language Principle, and which appears to be insoluble without a thorough reappraisal of the same. This issue, as the title of the book suggests, is that of self-awareness. The paradox runs as follows:

Let us suppose that we wish to gain an understanding of first-person cognitive development. How is it that a young human (for want of a better example; the extent to which the same phenomenon exists amongst other animals is debated inter alia) develops the ability to think about themselves? At which point in a child’s development do they first begin to gain a genuine sense of self-awareness? The Thought-Language Principle would tell us that, in order to truly understand this sense, we first must gain a thorough semantic understanding of the first-person pronoun.

The issue here appears to be a linguistic one but we should not get bogged down by linguistic considerations. The issue is semantic and has absolutely nothing to do with syntax; for this reason alone it should cross over language divides and should truly make no difference whether we speak of the English “I” or the French “moi” or the Hebrew “אני”. In all instances, Bermúdez argues, the meaning is identical. The first-person pronoun always refers to the individual who is utilising it (ignoring language-specific idiosyncratic uses like the condescending English “we” in “how are we feeling today?”, etc). It is only by gaining a thorough grounding in the meaning of this word that we are able to ground ourselves thoroughly in the meaning of the concept and – thus – in our understanding of its psychological development.

The problem here is that the usage of this term necessarily presupposes a thorough grasp of the concept itself. Nobody can use “I” as the subject of a sentence (again, this is a semantic and not a syntactic consideration; “subject” here does not refer to the word’s function relative to the verb, but relative to the narrator¹) and not have a genuine self-consciousness. But if they have a self-consciousness prior to developing an awareness of the word, then the Thought-Language Principle must be false. This is threshed out in considerably more detail by Bermúdez, who sets about establishing the nature of the paradox in several steps.

In brief, those steps are as follows (p24, 267):
1. The only way to analyse self-consciousness is by analysing the capacity to think first-person thoughts;
2. The only way to analyse a thought is through a semantic analysis of that thought’s expression [Thought-Language Principle];
3. First-person thoughts are always conveyed through the first-person pronoun (English “I”, “me”, etc);
4. Linguistic mastery of the first-person pronoun necessitates the ability to think first-person thoughts;
5. A non-circular appraisal of self-consciousness is possible;
6. Mastery of the semantics of the first-person pronoun meets the “Acquisition Constraints”: a rule that suggests that any psychologically real cognitive ability can have its development within an individual’s cognitive maturation explained.

Rules 5 and 6 are both desirable, but both are responsible for the paradoxical nature of this problem. Demanding that we require a non-circular account of an explainable phenomenon has rendered this conundrum seemingly insoluble for, according to the Thought-Language Principle, mastery of the first-person pronoun must precede the ability to think first-person thoughts, and yet first-person thoughts are demonstrably required prior to the possible formulation of a semantically correct usage of the first-person pronoun. As for straightening out the paradox, Bermúdez suggests a reformulation of the Thought-Language Principle itself.

Although he later rejects the Functionalist perspective, modifying it instead in order to allow for a particular ambiguity that exists in third-person representations of self-consciousness², this perspective has much going for it and constitutes – in my mind at least – a viable supplement to the rather hegemonic Thought-Language Principle. This principle basically states that, despite being apparantly circular, the cognitive development of self-awareness has a particular origin in a first-person appraisal of the world that lacks first-person recognition. An example might suffice to demonstrate what is meant by this.

Suppose that a sentient being was to encounter food. They might think to themselves, “food is before me”, and thus employ an (admittedly, objective) first-person pronoun. According to the functionalist perspective, there was a time in that being’s existence when it simply declared, “food!” This assertion, while a corollary of that being’s ability to see, does not necessary reflect their understanding of that same faculty. A being that never develops any sense of self-awareness is still capable of recognising objects outside of themselves (by which I would also include the parts of their body that, without a sense of self-consciousness, are merely appendages that move at whim). A human child, in other words, may initially perceive objects and only later develop the sense that they are seeing those objects. The advantage here is that this principle maintains the Thought-Language Principle as it is, and simply allows for a gradual development of cognitive faculties that resolves the paradox outlined by Bermúdez.

As mentioned, Bermúdez dislikes this principle as it is – partly because it does not allow for the subtle differences between different types of first-person experience, but chiefly because he is interested in thorough reformulation (and partial rejection) of the Thought-Language Principle itself. His book is an interesting one and, while much of the logical argument is thoroughly over my head, his conclusions appear to have been well-received. I would certainly be interested in reading more of Bermúdez’s argument but, for the time being, my money is still placed on the Thought-Language Principle as it presently exists according to the Functionalists. Like the man who seeks to read the barcode on his torch but realises that he cannot shine it upon itself, we are unfortunately limited in using our mind in our analysis of the same. I often wonder just how different an alien’s appreciation of our own cognitive development might be, and just how that might align with the conclusions that we, ourselves, have drawn. I have every faith that human consciousness is infinitely more complex than we could ever dream.

¹ This is explained, in the book, as being the difference between a sentence like “I have a pain in my tooth” and “My hair is blowing in the wind”. In the first instance, pain is an undeniable phenomenon and there is no room for doubt within the individual’s assertion that they feel it; the second instance, while it does contain a semantically correct usage of the first-person pronoun, also leaves inherent room for doubt. The first example uses the word subjectively, the second objectively – despite the fact that both are the syntactic subjects of their respective sentences.
² In brief, this is the ambiguity that exists between different senses of a phrase like “Simon realised that he was going to be the next person to have his name called out”. From one perspective, this might mean that Simon experienced a first-person thought that would have been formulated with the first-person pronoun. Another perspective, however, states that Simon experienced a third-person thought instead. Were Simon to be in a situation, for example, where alphabetical names are being read out from a list, he may know that “Simon” will be read out next but not know that the Simon to whom is being referred is he himself. It is important for our purposes that Simon is the one who is being called in order for this to represent a first-person thought that does not utilise the first-person pronoun. The Functionalist perspective, as outlined briefly above, does not take this anomaly into account; Bermúdez’s solution is effectively a reworking of the Functionalist perspective in order to accommodate this particular situation.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: