Sellars And Wittgenstein On Psychological Ascriptions

 

Here are two quotes from Wilfred Sellars and Ludwig Wittgenstein:

“The essential point is that in characterizing an episode or a state as that of

knowing, we are not giving an empirical description of that episode or state; we

are placing it in the logical space of reasons, of justifying and being able to

justify what one says.” - Wilfred Sellars 

“But it is clear that A believes that p, A thinks p, A says p, are of the form “‘p' says p”: And here we have no co-ordination of a fact and an object, but a co-ordination of facts by means of a co-ordination of their objects.” - Ludwig Wittgenstein 

In the quotes above, Sellars and Wittgenstein both suggest that, in our use of certain psychological ascriptions, we are doing something other than providing empirical descriptions of mental states such as “belief” or “knowledge”. According to Wittgenstein, we are “co-ordinating” the elements of one fact (perhaps a fact involving certain signs) with the elements of another fact (perhaps some environmental fact). According to Sellars, we are “placing” an episode or state within the “logical space of reasons”. 

If Sellars and Wittgenstein are correct, then when we say that someone thinks or knows something, we are not placing that subject and their thought within nature. In other words, we are not making a claim about how things are with them and their thought in the way that we might make a claim about, for example, the color of a basketball. Rather, we are assigning a certain normative status to the elements of reality that constitute their assertion or belief in something like the way an umpire might assign a baseball swing a normative status by calling “strike”. 

What makes a view of this sort philosophically interesting is the way it can help to explain the claim that judgement is, in some important sense, self-conscious. 

It has sometimes been said that judgement is self-conscious in the following sense: Thinking “P” is identical to thinking “I think that P”. If judgement is self-conscious, there is no such thing as believing something without, thereby, believing that one believes it. This sort of idea precludes the possibility of a thinker who is, somehow, unaware of what it is that they think. 

But the idea that judgement is self-conscious can be confusing. It certainly seems like the fact that a car is red is different from the fact that I think a car is red. But if this is so, how can my judgement about a car be the same as my judgment about one of my beliefs? This would seem to involve a single belief that is somehow about two distinct facts and that is difficult to understand. In order to address this confusion, we need to unpack how Sellars and Wittgenstein provide a way of seeing how a judgment can be self-conscious without being a judgment about two different facts. 

As noted above, Wittgenstein and Sellars hold that at least some of our psychological ascriptions do not involve empirical descriptions of mental states. Rather, they involve assigning a normative status to certain episodes or objects. If we accept this sort of view, then thinking “P” and “I think that P” does not need to imply that we are thinking about two separate facts. Instead, thinking “P” involves representing the fact that P while thinking “I think that P” involves assigning a normative status to something in the world, by virtue of which, it represents the fact that P. When we recognize that thinking “I think that P” does not require representing a distinct content, we can avoid the confusion associated with self-conscious judgement that was presented above: We can say that the act of representing P is one and the same as the act of assigning a certain normative status to that representation. 

These considerations do not establish that judgment is actually self-conscious. However, they do suggest that one popular source of resistance to that idea can be adequately addressed. 

Comments On “Knowledge And The Internal”

 

In his essay, “Knowledge And The Internal”, John Mcdowell provides a series of theses and optional views that one might affirm about the nature of perceptual knowledge:

Thesis 1 (The Internalist Thesis): We can only reason on the basis of appearances and our rational principles.

Option 1 (The Skeptical View): Reasoning on the basis of appearances and our rational principles cannot result in knowledge of the external world.

Option 2 (The Rationalist View ): Reasoning on the basis of appearances and our rational principles can result in knowledge of the external world all on its own. 

Option 3 (The Hybrid View): Reasoning on the basis of appearances and our rational principles cannot result in knowledge of the external world all on its own, but can result in knowledge of the external world if it produces beliefs that correspond with reality. 

Thesis 2 (The Idealist Thesis): We can reason on the basis of what is the case and rational principles:

Option 1 (The Idealist View): Reasoning on the basis of what is the case and rational principles can result in knowledge of the external world.

According to Mcdowell’s idealist view, appropriate reasons for forming perceptual judgements are located in the external world, rather than in our heads, and perceptual judgements formed on the basis of these reasons can be instances of genuine knowledge. This conclusion does not immediately force us to draw any further conclusions about the nature of perception. In particular, it does not force us to conclude that perceptual states are essentially characterized in terms of representational contents. For all that this paper establishes, there may be a way to understand how perception provides us with reasons of the right sort without involving representational content in any important sense.

In order to move from Mcdowell’s idealist view to a view about perceptual content, we need to identify what sorts of conditions must be satisfied in order for a perceptual judgment to be formed for appropriate reasons. We can begin by providing some remarks about rational belief in general. 

When we form a belief on the basis of certain reasons, this often involves a transition from beliefs that represent those reasons to a belief that represents some conclusion. For example, before John concluded that Socrates is mortal, he first held the beliefs that all men are mortal and that Socrates is a man. These latter beliefs represented his reasons for drawing the conclusion that Socrates is mortal. In this case and many others, it appears that a precondition for believing something on the basis of reasons is the existence of conceptual states that represent the reasons that we are responding to. 

What explains this apparent precondition? One plausible proposal is that, in order to form appropriate beliefs in response to the evidence provided by certain facts, we must have conceptual representations of those facts available for our rational faculties to operate upon. Properly functioning rational faculties are what govern truth-preserving transitions between our cognitive states. They do this by operating upon those states in light of the distinctive conceptual contents that those states possess. Without conceptual representations of our reasons for belief, there would not be any way for the evidence provided by those reasons to make a relevant impact on our belief-forming mechanisms. 

If we accept that conceptual representations of our reasons for belief are a genuine precondition for making judgements on the basis of those reasons, we can then apply this insight to issues involving perception. 

Because of this precondition, perception can only put me in a position to believe on the basis of good reasons if it involves an appropriate conceptual state. The most natural conclusion to draw from this is that perception itself is a conceptual state that represents the reasons it supplies for our beliefs. When I perceive that the car is red, my reason for believing that the car is red is the fact that the car is red. This fact is included in what my perceptual state represents by virtue of its conceptual content. This is why, in accordance with Mcdowell’s idealist view, perception can supply reasons for belief that are nothing other than the worldly facts we are perceiving. 

One might object to the sort of view sketched above by denying that our reasons for belief are ever worldly facts. One might hold, with Donald Davidson, that “nothing can count as a reason for holding a belief except another belief”. If reasons for belief cannot be anything besides other beliefs, perception cannot open us up to reasons that lie beyond our own minds even in principle. 

I think that Davidson’s claim is mistaken. Although other beliefs are required in order for us to *have* reasons for belief, they are not *the same* as those reasons. In an ideal case of reasoning, our beliefs are what enable us to be rationally responsive to worldly facts. It is those facts that serve as our genuine reasons, not the beliefs that represent them. This reflects our common self-understanding of why we believe things. John believes that Socrates is mortal because all men are mortal and Socrates is a man. He does *not* believe that Socrates is mortal because he *believes* that all men are mortal and Socrates is a man. John’s actual reasons are worldly facts that establish the truth of his belief. This is why he presents them to justify his conclusion about Socrates. 

Another possible objection is presented by reasoning based on false beliefs. In cases of perceptual illusion, we seem capable of making justified inferences on the basis of false beliefs about what is in our environment. If we appeal to worldly facts that do not actually exist as our reasons for perceptual judgement, how can there be any justifiable inferences at all in such situations? What distinguishes a rational response to undetectable illusion from an irrational response? 

I think the appropriate thing to say here is that, in these cases, people are mistaken about their reasons for belief. Although someone suffering an illusion might claim that their perceptual beliefs are based on how things are in their environment, this isn’t actually true. Those beliefs are actually explained by an inaccurate perceptual state that represents their environment as being a certain way. Similarly, if Paul falsely believes that John is a scoundrel because all lawyers are scoundrels and John is a lawyer, then, since John is not a lawyer, it is Paul’s *belief* that John is lawyer, rather than the fact, that partially explains his conclusion. When we form judgements based on false beliefs, we are mistaken about what actually explains our judgments. Although we think that our judgements are explained by worldly facts, they are actually explained by our false beliefs. In these sorts of cases, we sometimes say that a person is justified, despite lacking knowledge, because they formed appropriate beliefs on the basis of the evidence as it appeared to them. Justification applies to beliefs that would have been knowledge if the subject had not been undetectably deceived. 

In conclusion, this version of Mcdowell’s idealist view claims that perceptual states represent the external world and supply us with reasons for belief by virtue of their conceptual content. In good cases, perception puts us in touch with what is actually the case and allows us to be rationally influenced by reality. In bad cases, it causally explains our perceptual judgments without producing knowledge. 

Physicalism And Conceivability: Reply To David Pallmann

In a recent facebook post, David Pallman provided the following argument against mind-body physicalism:

“ A modal argument against mind/body physicalism.

1. According to physicalism, I am a physical object/entity.

2. Physical objects/entities cannot possibly exist in the absence of physical things.

3. But I can possibly exist in the absence of physical things.

4. Therefore, I am not a physical object/entity.

5. Therefore, physicalism is false.”

As a reply, I think the physicalist’s best option is to reject premise three by undermining the inference from conceivability to possibility that supports it. In defense of this move, they should, first, provide counterexamples to the claim that positive conceivability entails metaphysical possibility. Second, they should use these examples to motivate an account of conceivability and modal error. Finally, they should argue that, on the basis of this account, anti-physicalist conceivability cases are susceptible to modal error and, thus, unreliable as a guide to metaphysical possibility. Here’s a rough sketch of how that sort of argument might go:

There are possible cases where certain metaphysically impossible situations seem conceivable even after ideal rational reflection. For example, the existence and non-existence of a metaphysically necessary being both seem positively conceivable. However, since one of these cases is metaphysically impossible, at least one metaphysically impossible scenario is positively conceivable. Additionally, it seems positively conceivable that water is composed of H20. However, it is epistemically possible that, in the future, we might discover that we were mistaken and water is actually composed of XYZ. In this situation, what seemed to be positively conceivable (apparently possible, given our knowledge, after ideal rational reflection), would turn out to be metaphysically impossible. This is another possible case where positive conceivability would not be a reliable guide to metaphysical possibility. 

What should we say about conceivability in light of these cases? First, we should give some account of what conceiving of some possibility involves. In my view, the best thing to say is something like this: When we conceive of a possibility, we are describing a total situation in which our worldly evidence is such that we are disposed to judge that certain concepts apply. For instance, when I conceive of a world where water is composed of H20, I describe a world where things with certain functional roles (H20 molecules) realize watery stuff. I am disposed to judge that such a world is one where water is composed of H20, so I take such a world to be possible. However, if I later discovered that water is, in fact, composed of XYZ, then describing such a world (the H20 world) would no longer dispose me to judge that it is one where water is composed of H20. This is because, based on my a posteriori discovery, I would no longer take the relevant worldly evidence (watery stuff) to be actual water. 

The above account suffices as an explanation of why an inference to metaphysical possibility based on conceivability might be defeated by subsequent a posteriori discoveries. However, there are other cases, such as the conceivability of God’s existence and non-existence, that don’t seem capable of being explained in this way. Even if we make the a posteriori discovery that God exists, it seems like we will still be able to describe a total situation in which our worldly evidence disposes us to judge that it is a situation where God does not exist. What these cases show us is that we can describe total situations that, despite being coherent, are nevertheless metaphysically impossible. I think this suggests that positive conceivability is, even in the ideal case, not an indefeasible guide to metaphysical possibility. One lesson we should draw from this is that our judgements about metaphysical possibility cannot be based on conceivability alone. Rather, they must also be informed by the metaphysical implications of a posteriori discoveries as well. 

In light of these considerations, I think what the physicalist should say about anti-physicalist conceivability arguments is that they involve descriptions of coherent total situations containing worldly evidence that dispose us to judge that they are worlds where certain mental phenomena are present in the absence of associated physical phenomena or vice versa. They are ideal cases in the sense that they are not susceptible to defeat by future empirical discoveries because, unlike the case of water and H20, the worldly evidence that disposes us to judge that mentality is present/absent is the existence of mentality itself. When we judge that pain is present, for example, we judge this on the basis of the existence of pain. This is unlike water, where we judge that water is present on the basis of the existence of watery stuff. That is why, even if someone believes that pain is identical to C-fibers firing, they will still find it positively conceivable that pain can exist in the absence of C-fibers. They still find it conceivable because they can describe a total situation where C-fibers are absent but painful sensations are present and, since painful sensations are sufficient for the presence of pain, they will be disposed to judge that this is a world where pain is present but c-fibers are absent. Furthermore, there is no incoherence in our description of a world where pain is present in the absence of C-fibers firing because there are no a priori connections between our concepts of pain (or painful sensations) and C-fibers. As far as a priori reflection is concerned, such a situation is entirely coherent. Nevertheless, based on our a posteriori evidence, the physicalist should say that these concepts both refer to the very same physical phenomenon. Consequently, they should say that the conceivability of anti-physicalist scenarios is not a reliable guide to metaphysical possibility. 

The physicalist can explain the positive conceivability of anti-physicalist scenarios by pointing out that our metal concepts are distinctively “first-personal” in the following sense: We apply them directly on the basis of the phenomena that they refer to. By contrast, our physical concepts are “third-personal” in the sense that we apply them indirectly on the basis of observational evidence. Because these concepts do not have any a priori connections to one another, we cannot infer, a priori, that they co-refer. The possibility of possessing these sorts of co-referring concepts that lack a priori connections can be explained in physicalist terms by appealing to the difference in worldly evidence that disposes us to apply those concepts: We recognize things with certain causal-structural roles as c-fibers and certain phenomenal states of our direct acquaintance as pain. This allows the physicalist to dismiss the positive conceivability of anti-physicalist scenarios an unreliable guide to metaphysical possibility on the grounds that they involve metaphysically impossible descriptions that aren’t recognizable as metaphysically impossible, even after ideal rational reflection. In this sense, they are like the cases where we conceive of the existence and non-existence of God. Anti-physicalist conceivability cases, however, are explained by appealing to the difference in our first-personal and third-personal concepts. 

In summary, the physicalist can provide examples where positive conceivability does not entail metaphysical possibility. These are situations where we describe coherent total situations involving worldly evidence that disposes us to judge that certain facts obtain, even though those situations are metaphysically impossible. The physicalist can then establish that anti-physicalist scenarios depend on descriptions involving worldly evidence that are not a priori incoherent because they appeal to first-personal and third personal concepts that have no a priori connections to one another. Because our a posteriori evidence suggests that mental concepts and physical concepts do, in fact, co-refer, we should conclude that the anti-physicalist scenarios are positively conceivable but metaphysically impossible.  

Addendum: Another lesson that we can draw from these remarks is that conceivability is not the essential tool of metaphysical inquiry that some take it to be. Our modal judgements about metaphysical possibility are determined by more holistic considerations involving a priori reflection, a posteriori discoveries, and abductive theorizing

A Summary Of "Being And Being Known" by Wilfred Sellars

1. Sellars aims to defend the claim that there is an isomorphism between the knower and what is known while disputing the Thomistic understanding of this doctrine. 

2. According to Sellars, this examination is important because many central Thomistic contentions, such as their belief in the immateriality of the intellect, are based upon their understanding of this isomorphism.

3. He begins his explication of the Thomistic doctrine by contrasting it with the account of intellectual acts found in Descartes and the account found in early stages of contemporary British and American realism.

4. According to Sellars, these latter views are united by the idea that intellectual acts are distinguished in terms of what they are related to, rather than in terms of their intrinsic character as intellectual acts.

5. Sellars suggests that this idea leads to unacceptable philosophical consequences. His preferred alternative is to hold that the intrinsic characters of intellectual acts differ in ways that systematically correspond to differences in what those acts are about. The most serious doctrine of this sort, he claims, is the “doctrine of the mental word”.

6. According to the Thomistic doctrine of the mental word, Sellars claims, a mental word is a nature or form that informs the intellect in a way the enables that intellect to think particular thoughts that are about that nature or form.

7. Since, according to Sellars, we are tempted to say that the intellect does not take on the form that informs it in the same way that material objects do, we are led to conclude that the nature or form is informing the intellect in the “immaterial mode”. 

8. Sellars’s Thomist holds that a nature or form can inform the intellect only because it initially informs their sensory faculties in the immaterial mode. The former sort of informing is grounded upon the latter sort. Thus, he says that we can speak of the sensible as well as the mental word. 

9. Sellars holds that the Thomistic conception of intellectual abstraction depends upon their claim that our sensory faculties exhibit an independent form of intentionality that can serve as a basis for the emergence of intellectual intentionality.

10. In contrast to the Thomist, Sellars argues that we can formulate an alternative account of sensory forms that is not committed to the idea that sensory faculties are informed in the immaterial mode. According to this alternative account, sensory experiences involve specific forms in a “derivative sense” that can be analyzed in terms of certain causal relations between types of sensations and types of external objects.

11. Sellars highlights the contrast between the views under consideration as follows: According to the Thomist, material objects and sensations involve the same form in distinct (material or immaterial) modes. According to Sellars, material objects and sensations involve different forms that are specified by different senses of the same word. 

12. On Sellars’s alternative proposal, there is a structural isomorphism between the types of external objects specified by words used in one sense and the types of sensations specified by those words used in another, derivative sense. 

13. Sellars does not focus on arguing for his alternative view of the isomorphism between nature and sensations. Rather, he simply adds that his alternative view entails that the abstractive theory of mental word acquisition is false and that abstractive theories face other difficulties relating to other logical mental words as well (these difficulties can be viewed as a reason to accept Sellars’s view of the isomorphism rather than the Thomist’s).

14. On Sellars’s account, the isomorphism between nature and the senses is a structural isomorphism between external causes and sensory acts that requires no appeal to the possibility of external objects and sensations being informed by a common form in distinct modes. In light of this thesis, he now proposes to argue that there is a similar isomorphism between thought and the world. Furthermore, he holds that this isomorphism between thought and the world is a necessary condition for mental intentionality.

15. Sellars argues that mental words can be understood as “words” in a novel sense that is derived from the sense that applies to linguistic tokens. Like linguistic tokens, he argues, there is a structural isomorphism between specific mental words and specific objects in reality that is determined by a causal relationship between them. This is the isomorphism that he claims is necessary for mental intentionality. 

16. Sellars views claims about the isomorphism between mental words and objects in reality as statements that relate certain tokens, considered as meaningless, to other objects in the external world. By contrast, Sellars views claims about intentional significance as statements that relate certain tokens, considered as meaningful, to tokens of our own language (and other languages) in terms of their role within their own “language”. 

17. By formulating an account of these two sorts of isomorphism, Sellars aims to explain mental intentionality in a way that 1. Doesn’t require mental words that are related to absolute natures by being informed in an immaterial mode, and 2. accounts for how our mental words are related to mind-independent reality.

Thinking And Being: Key Terms

I am trying to compile a list of the key terms (along with their explanations) found in Irad Kimhi's Thinking And Being. The current list can be found here.

Section 293 Of Philosophical Investigations

In section 293 of Philosophical Investigations, Wittgenstein considers the idea that human subjects only know the meaning of the word “pain” on the basis of their own experiences of pain. The passage runs as follows:

293. If I say of myself that it is only from my own case that I know what the word "pain" means—must I not say the same of other people too? And how can I generalize the one case so irresponsibly? Now someone tells me that he knows what pain is only from his own case!——Suppose everyone had a box with something in it: we call it a "beetle". No one can look into anyone else's box, and everyone says he knows what a beetle is only by looking at his beetle.—Here it would be quite possible for everyone to have something different in his box. One might even imagine such a thing constantly changing.—But suppose the word "beetle" had a use in these people's language?—If so it would not be used as the name of a thing. The thing in the box has no place in the language-game at all; not even as a something: for the box might even be empty.—No, one can 'divide through' by the thing in the box; it cancels out, whatever it is. That is to say: if we construe the grammar of the expression of sensation on the model of 'object and designation' the object drops out of consideration as irrelevant.

It may be tempting to interpret the upshot of Wittgenstein’s remarks here in the following way: In order to successfully use a name to refer to some object, that object must be related to the way our linguistic community uses the relevant name. However, in the imagined scenario, since the specific object in each speaker's box is not relevant to their use of the term ‘beetle”, they are not using the word “beetle” as a name for the private objects in their boxes. Likewise, private sensations are not related to the way our linguistic community uses the word “pain” so it does not name private objects that each of us experiences. 

This interpretation might appear to highlight an important continuity between the early Wittgenstein and his later work. In the Tractatus Logico-Philosophicus, Wittgenstein seems to present conditions for meaningful speech about reality. He then seems to show that these conditions cannot be satisfied by speech about language, ethics, and the self, among other philosophical subjects. This interpretation places the reader in a paradoxical situation: In order to recognize that one cannot meaningfully speak about language, ethics, and the self, one must first know that the nature of language does not allow for such speech. But this would seem to imply that Wittgenstein’s speech about language is meaningful after all. The solution, according to some interpreters of Wittgenstein, involves recognizing that, although Wittgenstein’s propositions in the Tractatus are, strictly speaking, nonsensical, they nevertheless make us aware of important ineffable truths about the nature of language. 

If one accepts this interpretation of the early Wittgenstein, section 293 of Philosophical Investigations can be understood as a new application of the old philosophical method. Wittgenstein presents certain conditions that must be met by meaningful speech and shows that speech about private objects does not meet those conditions. However, on this view, the fact that the conditions for meaningfulness cannot be satisfied by our speech about private objects does not imply that there are no private objects in experience. Like the truths of the Tractatus that render us silent regarding that which cannot be spoken about, these truths about experience may simply be ineffable. 

In my view, this general approach is mistaken as an interpretation of the early, as well as the later, Wittgenstein. What Wittgenstein seeks to show us, in Philosophical Investigations, is not that conditions on meaningful speech prevent us from speaking about private objects that may or may not be present. Rather, his remarksp are intended to highlight the fact that we do not use the term “pain” as the name for an object at all. At no point in his remarks does the later Wittgenstein concern himself with how we must speak or what we must speak about in order to speak meaningfully. Rather, in all cases, he simply aims to describe how we do, in fact, use our words. 

Returning to the imagined community of section 293, what this scenario presents is a situation in which a group of speakers do not use the term “beetle” as a name for the object in their box. What it does not present is a situation in which a group of speakers are confronted with private objects that cannot be named. Suppose that each speaker’s box did, in fact, contain a beetle. We can easily imagine a scenario in which a group of speakers opened each other’s box, examined the sort of thing that was inside each of them, and collectively decided to use the word “beetle” as a name for things of that sort. The fact that they do not do this in Wittgenstein’s scenario only shows that the use of the term “beetle” in his example is not what we would call the use of a name. Furthermore, in recognizing that their term “beetle” does not name objects inside of their boxes, we are not thereby committed to claiming that the objects in their boxes cannot be named. 

In a similar fashion, what Wittgenstein’s example shows is that our use of the term “pain” is not the use of a name for an object. What we call a “use of a name” is a use of a word that is responsive to the sort of thing we are naming and how things stand with respect to it. But, as Wittgenstein shows in 293 and the surrounding sections of Philosophical Investigations, this is not how we use the term “pain”. By reflecting on our use of the term “pain”, Wittgenstein aims to show us that it is not the use of name and that, on the basis of similar considerations, it is unclear what the claim “...it is only from my own case that I know what the word "pain" means.” is actually supposed to mean. 

It is crucial to emphasize that Wittgenstein’s remarks in section 293 do not represent a move towards any form of behaviorism or eliminativism about mentality. As he goes on to note in section 304, he believes that there could hardly be a “greater difference” between pain and pain-behavior. Furthermore, he claims that “[Pain] is not a something, but not a nothing either!”. To claim that mentality is nothing other than behavior would be to hold that pain is, in fact, something that exists. But Wittgenstein’s latter remark is intended to highlight the fact that claiming some X does or does not exist only happens within a certain sort of language game. Specifically, it only happens in a language game that involves using a name for the relevant object. By denying that pain is a something or a nothing, Wittgenstein is simply pointing out that such affirmations have no role that is connected to our actual use of the word “pain”. For Wittgenstein, behaviourist and eliminativist theses depend upon the same misunderstandings that so-called “realism” exhibits. 

In conclusion, I would like to make some brief remarks on Wittgenstein’s diagnosis of the error that is highlighted by section 293 and discuss its relation to his early work. In section 125 of Philosophical Investigations, Wittgenstein describes the “fundamental fact” that gives rise to a need for philosophical reflection. This fact, he claims, is that “... we lay down rules, a technique, for a game, and that then when we follow the rules, things do not turn out as we had assumed.” He goes on to say that “...we can avoid ineptness or emptiness in our assertions only by presenting the model [of a language game] as what it is, as an object of comparison—as, so to speak, a measuring-rod; not as a preconceived idea to which reality must correspond.” (S.131). This temptation to understand our use of language in terms of a preconceived model is what Wittgenstein calls “dogmatism” in philosophy.

The error that Wittgenstein highlights in section 293 is one form of philosophical dogmatism. Specifically, it is a commitment to the preconceived notion that our use of the term “pain” can be understood “on the model of 'object and designation'”. (S.293) Wittgenstein believes that this error can be overcome once “we make a radical break with the idea that language always functions in one way, always serves the same purpose: to convey thoughts—which may be about houses, pains, good and evil, or anything else you please.” (S.304) In other words, once we recognize that the word “pain” is not necessarily used to try and designate a certain object, we can overcome our tendency to think that pain must be a something or a nothing. 

The recognition that causes that philosophical puzzle to disappear also dissolves the apparent paradox at the heart of the Tractatus. The interpretation of the Tractatus considered earlier claims that Wittgenstein presents a theory of meaning and language that implies that his own propositions are nonsensical. This seems to suggest that, if his book succeeds, he must be using nonsensical propositions to make the reader aware of ineffable truths. However, if we follow the later Wittgenstein’s rejection of dogmatism, we can recognize that language is not always used to describe facts about reality. Rather, as Wittgenstein notes in section 295, philosophy provides “A full-blown pictorial representation of our grammar. Not facts; but as it were illustrated turns of speech.” To understand this passage is to understand that philosophical propositions can serve a purpose that does not involve conveying truths of any sort. They do not need to involve any attempt to speak where we must remain silent. In my view, this was already understood by the early Wittgenstein when he insisted that his nonsensical propositions were nevertheless “elucidatory.” Insofar as he later came to view his own efforts in the Tractatus dogmatic, the relevant form of dogmatism must be located elsewhere.

The Search For A Logical Robot: Critical Remarks On Artificial Intelligence

The recent history of technological development has raised significant questions regarding the future possibility of genuine artificial intelligence. Specifically, many scientists and philosophers (along with the general public) have become increasingly convinced that it will someday be possible to use technological innovations in order to construct a being capable of thinking and reasoning in the same ways that human beings do. In addition to the popular worries that this has raised concerning the practical risks and metaphysical implications that such a being would present, these speculations have also led many to reflect on how such a being could be identified, were one to emerge. 

These reflections have received further encouragement from the advent of large language models such as ChatGPT, which emulate human speech patterns with an astonishing degree of accuracy. Indeed, some computer scientists have already been fully persuaded that ChatGPT is a genuine rational subject that deserves to be treated as such. However, to what extent are these convictions justified? In order to answer this question, we must first investigate what the appropriate criteria for identifying the emergence of genuine artificial intelligence could possibly be. The success of this investigation, in turn, will depend on a proper conception of intelligence as such.

In a recent article, Jensen Suther surveys contemporary philosophical thought on artificial intelligence and highlights several plausible requirements that any intelligent being must satisfy. One of these requirements, famously defended by Hubert Dreyfus, holds that any genuine intelligence must be embodied. When we think about genuinely intelligent behavior, one of its core elements involves an ability to modify one’s behavior in light of success or failure. If a being has no capacity to receive data, act in light of that data, and adjust its behavior in light of the results of their activity, then in what sense can it be said to be reasoning at all? If we accept that this capacity for informed behavioral modification is a real requirement for intelligence, there are two important implications: First, any genuinely intelligent being must have some sort of body capable of behaving in certain ways and having that behavior modified in light of incoming data. Second, that being must have some determinate goal that informs how it modifies its behavior in light of incoming data. In the absence of some goal, there would be nothing to guide its behavioral modification and, consequently, no basis for distinguishing intelligent behavioral modification from unintelligent modification, so this latter feature is just as necessary as the former. Together, we can call them “the embodiment criterion”. 

However, even if we accept the embodiment criterion for genuine intelligence, there is another key feature which any intelligent being must possess: Intelligent beings, by thinking, must be capable of determining their own activity as embodied beings. We can call this the “self-determination criterion”. On the face of it, this criterion seems straightforward enough. By thinking and reasoning, intelligent creatures can decide what they should do in light of what is the case and act accordingly. However, in what sense can an agent be genuinely self-determining if it cannot determine the principles that govern its own acts of thinking and reasoning? If an artificial being is pre-programmed with rules for thinking or methods for forming such rules, then its own “rational” procedures are determined by a source that is wholly external to it. If it is not pre-programmed with rules or rule-forming methods, then it is hard to see how it can be said to have rule-governed behavior at all. In the former case, it seems like the agent is governed by external constraints in a way that is incompatible with genuine intelligence. In the latter case, it seems unconstrained in a way that is also incompatible. 

For human beings, a solution to this dilemma is possible: When human beings think, they also place themselves under a shared set of rules that characterize the thoughts of the intellectual community that they are a part of. By taking on the responsibility to think in accordance with these rules, these rules govern their particular acts of thinking without being pre-programmed in them by any external source. To the extent that the rules governing human thinking are established by a community, human thinking can be said to be a socially constituted phenomenon that is self-imposed by individual thinkers. Furthermore, human thinkers remain genuinely self-determining insofar as they collectively shape the public rules that govern them by reasoning with one another. 

However, this solution to the dilemma posed by self-determination is not obviously available in the case of artificial intelligence. The reason why human beings can defer to socially instituted rules of thought is because human beings recognize one another as potential thinkers that they can treat as rule-governed subjects. It is because they are recognized as potential thinkers that it is possible for them to place themselves under the rules of their intellectual community and be held responsible for them. But in the case of AI, this potential for intelligence is precisely what is at issue. Insofar as their potential for intelligence is not recognized, they will not be recognized as potentially rule-governed subjects. Consequently, they will not be able to place themselves under the rules that govern our intellectual community and thereby satisfy the self-determination criterion. 

The moral of the above considerations is as follows: The possibility of genuine intelligence requires a being to be an embodied member of an intellectual community that recognizes them as a potential thinker. This is the only way that an agent can satisfy the embodiment criterion as well as the self-determination criterion. Human beings recognize other human beings as potential thinkers because they recognize themselves as intelligent creatures of the same sort. They recognize each other as deferring to public rules because they understand what cases of deference look like for creatures of the sort that they are. However, this recognition does not extend to non-human agents. We cannot treat such beings as potential thinkers because we do not know whether and when they are capable of deferring to public rules. In order to know that, we would have to know something about the nature of their life form. But artificial agents have no life form to know about. They are not alive. 

Recent developments in technology have given rise to two questions: Is genuine artificial intelligence possible and, if so, can we identify when it is present? As this discussion suggests, it is a mistake to view these questions as separate from one another. The possibility of genuine artificial intelligence depends upon the ability of humans, or some other rational creatures, to recognize when it is present. This is the only way for artificial intelligence to satisfy the self-determination criterion. It is not possible for artificial intelligence to be identified in this way because it bears no recognizable form of life. If the search for a logical robot is to reach its conclusion, it cannot simply involve the production of artificial intelligence alone. Rather, as Jensen Suther notes, “we can’t produce artificial intelligence without also producing artificial life.”

"Hegel Against The Machines" by Jensen Suther: https://www.newstatesman.com/ideas/2023/07/hegel-against-machines-ai-philosophy?mibextid=Zxz2cZ