10: Concepts and Content -- Introduction to Philosophy- Philosophy of Mind

Introduction

The problem of intentionality is the problem of how some entities can be “about” something. That is, sentences, thoughts, or concepts, among others, display intentionality in that they are about something else; they are said to be a representation of something. The notion of intentionality can be traced back at least as far as Aristotle (384 BCE-322 BCE), though the German philosopher Franz Brentano (1838-1917) is generally credited with introducing the notion to contemporary philosophy in the late nineteenth century. Brentano’s oft-quoted remark is that “Every mental phenomenon is characterized by … the intentional (or mental) inexistence of an object” and “reference to a content, direction toward an object.” In other words, “Every mental phenomenon includes something as object within itself, although they do not all do so in the same way. In presentation something is presented, in judgement something is affirmed or denied, in love loved, in hate hated, in desire desired and so on” (Brentano [1874] 1995, 68). The usual way to frame the problem of intentionality is in terms of the notion of meaning or content. What is the status of the meaning of a sentence over and above its formal and syntactic aspects? What makes it the case that a particular proposition has the content that it does? Is content only dependent upon mind-internal properties? Or must we make use of mind-external factors such as the context of the utterance or the speaker’s social history in order to determine the content? Those who argue that the relevant and scientifically interesting properties that are involved in content are overwhelmingly, though not entirely, within the mind are referred to as internalists. On the other hand, externalists argue that there is something more to content than merely mind-internal events and their happenstance connection to the world: externalists insist that the meanings of our words (or sentences, or the contents of our thoughts, etc.) depend on some deep metaphysical (perhaps causal) connection between the mind and other worldly objects that are independent of the mind.

Externalists argue that a theory of content needs to provide an account of the relation between linguistic expressions and what may be called things in the world. In other words, the claim is that in order to explain content we must provide an account of the relation between linguistic expressions and the things that they can be used to talk about. Or as Colin McGinn puts it: “[E]xternalism supposes there to be a deep connection between states of mind and conditions in the nonmental world. Is the mind fundamentally autonomous with respect to the world, or does the world enter into the very nature of the mind?” (McGinn 1989, 1) He remarks further that according to externalism, “The environment is thus held to be constitutive of the very nature of mental states, determining what they are.” McGinn argues that internalism “insists upon … drawing a sharp line between mind and world; but the externalist holds that the mind is penetrated by the world, configured by it” (1989, 3). That said, however, we will see that the main force and substance of the internalist position is not exactly a mirror image or a negation of the externalist position, for internalism only denies that there is a deep metaphysical relation between the things in the world and linguistic expressions. That is, internalists dispute the externalist claim that the relations between linguistic expressions and the things in the world are desirable or even tractable in an explanatory theory of content.

This chapter is structured as follows. First I discuss the nature of concepts. I then discuss externalism and the way in which it explains the nature of concepts and their content. I discuss one of the major thought experiments that have motivated many philosophers to adopt the externalist position. I then discuss the internalist position, which not only provides objections to the main claims of externalism but also provides its own positive account of concepts and their content.

What is a Concept?

A concept is generally understood in the philosophy of mind to refer to a constituent of thought. Consider the proposition “John thinks that the book is on the table.” Following Bertrand Russell (1872-1970), philosophers speak of propositional attitudes, which include beliefs, desires, hopes, fears, expectations, and any other attitude that involves a proposition. A propositional attitude of the form “X thinks that P” has two parts. The first constitutes the verb of the psychological-state description and contains information about the agent and the agent’s psychological state. That is, the first part gives information about who takes which propositional attitude (for example, “John thinks”). The second part completes the description by revealing what the proposition is, or what the attitude refers to (for example, “that the book is on the table”). Concepts are said to be the constituents of the propositions expressed by propositional attitudes (in this case the concepts are, roughly, “book,” “table,” and “on”). Propositional attitudes and thus concepts are used in folk psychological (or intentional) explanations of behavior.

A simple example of the explanations and predictions that folk psychology allows is as follows. Suppose that we wish to explain why Leila chose to take her umbrella with her when she departed today. We can make use of some propositional attitudes and certain laws of folk psychology to formulate such an explanation. For instance, Leila believes that it will rain today (perhaps she heard the weather forecast on the radio), Leila believes that using an umbrella will help her seek shelter from rain, Leila desires not to get wet today. Therefore, since, all things being equal, humans act in accordance with their beliefs and desires, we can explain why Leila took her umbrella with her today. In other words, she took the umbrella because she believed X and desired Y, and believed that by doing Z she can bring it about that Y (notice that this is a counterfactual, so that if she did not believe that X and did not desire that Y then she would not do Z). Since beliefs and desires are integral parts of human thought, and since beliefs, say, are thought of as expressing propositions, the central role of concepts in philosophy and psychology is clear. That is, since the constituents of propositions are concepts, the need to explain the nature of concepts is inseparable from a theory of how the mind works.

Concepts are clearly shared between different people, and the question of what it is that is shared can be understood as the question of what the nature of content is. That is, if both John and Leila think that P, then they both share the content inherent in the concepts of the proposition P. What this claim amounts to is cashed out in very different ways by externalists and internalists, especially in terms of what explanatory role content is supposed to play. Externalists are mostly interested in concepts insofar as they figure in explanations of behavior (linguistic or other), whereas internalists are interested in concepts insofar as they serve as the meanings of linguistic items. Thus, a focus of internalist semantics is the underlying mechanisms of conceptual structure in virtue of which language production and comprehension is made possible. Let us now see what each claim amounts to.

Externalist Explanations of Content

Since the 1970’s externalism has become a widely held position in the philosophy of mind. The classic arguments for externalism are found in Hilary Putnam’s (1926-2016) “The Meaning of ‘Meaning’”, Tyler Burge’s “Individualism and the Mental”, and Saul Kripke’s Naming and Necessity. Putnam argues that “a better philosophy and a better science of language” must encompass the “social dimension of cognition” and the “contribution of the environment, other people, and the world” to semantics (Putnam 1975, 49). Burge argues against any theory about the mind in which “the mental natures of all a person’s or animal’s mental states (and events) are such that there is no necessary or deep individuative relation between the individual’s being in states of those kinds and the nature of the individual’s physical or social environment” (Burge 1986, 3-4).

The Twin Earth thought experiment of Putnam is the most famous argument in favor of externalism; it claims to show that two subjects can have identical internal psychological mental states but that the content of these states can be different due to particular variations in the environment. Putnam asks us to imagine a world (Twin-Earth) in which water is not composed of H₂O like it is on our world but is rather composed of XYZ. When a person (call him Oscar) says “water” on Earth the word refers to H₂O, but when a different person (call him Twin-Oscar) says “water” in a different place (on Twin-Earth) the word refers to XYZ. This seems intuitively clear; the word “water” refers to what the word is about in that particular environment (so when Oscar utters “water” that word is about H₂O in his environment). Putnam asks what would happen if Oscar is transported to Twin-Earth. Would the word “water” uttered by Oscar on Twin-Earth now refer to H₂O or XYZ? Notice that the thought experiment legislates that the only change that takes place when Oscar is transported from Earth to Twin-Earth is the change in his environment (i.e., all of his psychological states remain unchanged). Now, Putnam reasons that if knowing the meaning of a term is just a matter of being in a certain psychological state, then “water” on Twin-Earth when uttered by Oscar should refer to H₂O and not to XYZ as we might expect. This is because Oscar’s psychological state was fixed on Earth, and if the psychological state fixes the reference then “water” refers to H₂O regardless of the environment the subject is in (Putnam 1975).

Another way to put the matter is as follows: when Twin-Oscar on Twin-Earth says “water” whilst pointing to a lake that is entirely composed of XYZ, as all watery things are composed of on Twin-Earth, “water” refers to XYZ and not to H₂O. But, Putnam’s argument claims, if knowing the meaning of a term is just a matter of being in a certain psychological state then “water” uttered on Twin-Earth by Oscar transported from Earth cannot mean XYZ and must mean H₂O. Something seems to be wrong here. If two people utter the same word in the same environment we expect that word to refer to the same thing. Thus, if we want to hold on to the claim that the meaning of a term determines its reference or extension then, the argument claims, we must concede that, as Putnam famously put it, “Cut the pie any way you like, ‘meanings’ just ain’t in the head!” (1975, 144) That is, the claim is that mind-internal properties on their own cannot fix the meanings of words or what their reference is.

Putnam’s argument is directed at the meanings of words, of course, but it was soon noticed by Colin McGinn, Tyler Burge, and others that the same argument also applies to the contents of our propositional attitudes, hence to the contents of our thoughts. The main claim of externalism, then, is that even though thoughts are said to be inside a person’s head, the content of these thoughts supervene^[1] on external factors in the environment of the person who has them. Thus, as Ben-Menahem remarks in regard to one of Putnam’s examples, “to speak of coffee tables it does not suffice for us merely to have the concept of a coffee table, but we must be in contact with actual coffee tables” (Ben-Menahem 2005, 10; emphasis in original).

Internalist Semantics

Now, it could be objected that externalism has to be right: How could content not depend on the outside world? Surely the meaning of the word “elephant” cannot be due to only mind-internal properties. The word is about elephants, it could be argued, which are in the mind-external world, not inside the mind. As we will now see, internalists argue that there are good reasons to question the externalist claim that concepts are connected to the world in the way in which externalists claim they are. In other words, internalism does not deny the link to the outside world but rather has a different explanation of how our mind generates and interprets the content of our concepts. Internalism argues that, for the purposes of scientific inquiry into language and mind, the internal properties of the human mind are the most relevant and fruitful subject matter. Thus construed, internalism is not so much a solution to the issues that externalists grapple with. Rather, as we will see below, internalism is a different research program, and so there is a difference in the sorts of questions externalism and internalism attempt to answer.

A succinct definition of internalism is provided by Wolfram Hinzen: “Internalism is an explanatory strategy that makes the internal structure and constitution of the organism a basis for the investigation of its external function and the ways in which it is embedded in an environment” (Hinzen 2006, 139). Internalism studies the internal structure and mechanisms of an organism; the external environment comes into the picture when the internal processes are ascribed content by the theorist, thus explaining how the internal mechanisms constitute a cognitive process in a particular environment. Such content ascriptions, claim internalists, vary with the theorist’s interests and aims, but the content and its ascription are not an essential part of the theory itself. So, for example, the mechanism that detects vertical lines in the visual input will be in the scope of an internalist theory, but not the representational content ascribed to the output of this mechanism. The latter could be any number of things (the vertical line could represent the edge of a building or be part of a larger representation of a human face) but its underlying mechanism remains unchanged.

In other words, as discussed by Frances Egan in work spanning the last few decades, the internalist claim is that the computational characterization of an internal mechanism abstracts away from specific content ascriptions. So one can separate the content of, say, a visual state, from the computational machinery in virtue of which that content is made possible. For example, a particular mechanism may receive as input a certain set of parameters that require computation. The theorist then examines the output of these computations. One can imagine a mechanism that is embedded in the visual system being ascribed visual contents befitting the theory of vision. However, that same mechanism can be embedded in the auditory system and thus be ascribed different (auditory) contents that are defined not in terms of visual properties but rather in terms of acoustical properties. In other words, there is nothing inherent to the internalist computations performed that makes them visual or auditory. As Egan has discussed, there is an underlying set of computations that are required for both visual and auditory processing. The label we give to the output of such computations (the content we ascribe to them) depends on where the input to the mechanism came from. If the input is visual then the theorist will ascribe the output of the mechanism a visual content. But nothing in the internal mechanism itself tells us that. Internalism studies the internal mechanism itself, which remains unchanged regardless of whether it happens to be embedded or used by, say, the visual system or the auditory system.

Let us unpack these claims. First, note that the externalist claim that states of individual organisms cannot be understood in complete isolation from the environment in which they happen to be is not in dispute. The argument is about whether what happens in the environment should be part of what the theory is supposed to explain. So what is the problem that internalists see with the externalists’ relation between words and the things words are used to talk about? In a classic paper that formed one of the foundations of internalist semantics, Jerrold Katz (1932-2002) and Jerry Fodor (1935-2017) discussed this issue (though they did not frame it in terms of internalism versus externalism). Katz and Fodor ask the reader to compare the following three sentences:

(1) “Should we take junior back to the zoo?”

(2) “Should we take the lion back to the zoo?”

(3) “Should we take the bus back to the zoo?”

They then remark that information that figures in the choice of the correct interpretation for each of these sentences includes the fact that, say, lions, but not children and busses, are often kept in cages. That is, unlike (2), the meaning of (1) cannot be that we should take a living being back to the zoo and put them in a cage. (1) means that we should take a child and show them the animals around the zoo. (3), on the other hand, has neither of these interpretations. (3) cannot mean that we should take the bus to the zoo and put it in a cage, nor can it mean that we should take the bus and show it the animals around the zoo. In order to decipher these meanings one needs to know certain facts about the world; these facts are not semantic or grammatical facts (Katz and Fodor 1963).

After listing a handful of other examples of what information is needed for interpretation, Katz and Fodor note that the reader will find it easy to construct an ambiguous sentence whose resolution requires the prior representation of practically any relevant item of information about the world. This is because, in order to resolve a great deal of ambiguous sentences, one needs to have certain facts about the world without which certain interpretations of sentences are unavailable. For example, consider the sentence “I saw the man with the binoculars.” If one didn’t know what binoculars were, then the only interpretation available would be that a person saw a man holding an object that is called “binoculars.” However, once one’s world knowledge is expanded to include facts about binoculars (namely, that they are used to see objects far away) then further interpretations become available and then the sentence becomes ambiguous. That is, the sentence then has the additional interpretation that “I saw a man and I used binoculars to see that man.” Katz and Fodor’s claim is that in order to disambiguate such sentences one needs to know things that are not purely semantic (what binoculars are used for is a fact about the world and not a grammatical fact). The problem is that a theory of meaning that aims to include all relevant information that is needed in order to disambiguate sentences and determine the correct interpretation will run into great difficulties, for such a theory cannot predict in advance what sort of information will be needed for sentence interpretation.

The upshot is that a theory that insists (as externalism does) on including the mind’s relations to the external world in a theory of language cannot hope to find reliable relations of the sort described above (let alone systematizing them into a fruitful explanatory theory). However, in regard to the underlying mechanisms of the mind in virtue of which the generation and interpretation of content is made possible, internalists claim that a fruitful theory is possible. The most famous proponent of internalist semantics is Noam Chomsky, and his work stands in stark contrast to the externalist semantics of Hilary Putnam or Donald Davidson (1917-2003). The recent work of Paul Pietroski (2008, 2010) is an excellent example of internalist semantics. Pietroski construes meaning in terms of blueprints that are used by the language faculty in the mind in order to construct concepts. This is an attempt to explain the underlying mental mechanisms in virtue of which we can generate and interpret thought contents. The meaning of a word in internalist semantics is cashed out not in terms of the word’s relation to the outside world but rather in terms of the word’s internal role in the mind’s construction of the concept that has the required content. Note the difference here. Internalist semantics studies the mechanisms in the mind that build concepts. Once these concepts are generated they are transferred from the language faculty to the mind-internal systems of thought and to the articulatory-perceptual system.^[2] These systems then make use of concepts for various ends such as thinking and talking about the world.

In other words, the internalist claim is that the mind has certain mechanisms that include instructions to build concepts, which then provide the inputs to other (mind-internal) systems that enter into various human actions, one of which is communication. Internalist semantics, then, concerns the nature of the computational mechanisms of the language faculty and their relation to the systems of thought; it concerns not the concepts themselves but the mechanisms that fetch, build, and combine concepts within the mind. That is, internalism is concerned with the mind-internal mechanisms of concept creation. This is of course one step removed from what externalist semantics studies, which are the concepts themselves, their role in language use, and their relation to the speaker’s environment.

Conclusion

The difference between the externalist and internalist position in regard to mental content is a difference in the sort of questions that each attempts to answer. It is a difference in the way in which each construes the role that content plays in the explanation of language and mind. The argument in favor of either approach to the science and philosophy of language and mind is of course not a knockdown argument, nor is it a guarantee that one side will turn out to be the correct path. As Gabriel Segal correctly remarks, “The point is that we should not expect to discover too much from the armchair. Discovering the true nature of content should be a scientific enterprise (whether we also call it ‘philosophical’ or not)” (Segal 2000, 20).

References

Ben-Menahem, Yemima, ed. 2005. Hilary Putnam. Cambridge: Cambridge University Press.

Brentano, Franz. (1874) 1995. Psychology from an Empirical Standpoint. London: Routledge.

Burge, Tyler. 1986. “Individualism and Psychology.” The Philosophical Review 95(1).

Hinzen, Wolfram. 2006. “Internalism about Truth.” Mind & Society 5.

Katz, Jerrold and Jerry A. Fodor. 1963. “The Structure of a Semantic Theory.” Language 39(2): 170-210.

McGinn, Colin. 1989. Mental Content. Oxford: Blackwell.

Pietroski, Paul M. 2008. “Minimalist Meaning, Internalist Interpretation.” Biolinguistics 2(4) (2008): 317-341.

Pietroski, Paul M. 2010. “Concepts, Meanings and Truth: First Nature, Second Nature and Hard Work.” Mind & Language 25(3): 274-278.

Putnam, Hilary. 1975. “The Meaning of ‘Meaning.’” Minnesota Studies in the Philosophy of Science 7.

Segal, Gabriel M. 2000. A Slim Book about Narrow Content. Cambridge, MA: MIT Press.

Further Reading