Knowledge and Belief

According to the tradition stemming from Plato, knowledge is a special type of belief. In order to know that p one has to believe that p and meet various further conditions on that belief—that the belief be true, that it be justified, that it not be accidentally true, etc. The object of knowledge is a proposition of a certain sort, and one stands in the relation of belief to that proposition; in the jargon, knowing that p is a “propositional attitude” with belief as its foundation. We can call this the “doxastic theory of knowledge”: knowledge is belief restricted and qualified.

            But let us review some linguistic facts that cast doubt on this venerable theory. Consider the sentence “John knows that London is the capital of England”. We can paraphrase that sentence by “John knows the fact that London is the capital of England”, but we can’t paraphrase it by “John knows the proposition that London is the capital of England”. The latter sentence sounds like it means that John is acquainted with a certain proposition, if it means anything, but this isn’t what the original sentence means. On the other hand, the sentence “John believes that London is the capital of England” can be paraphrased by the sentence “John believes the proposition that London is the capital of England”, but not by the sentence “John believes the fact that London is the capital of England”. The latter sentence sounds like it means that John believes a certain fact, if it means anything, but that is not what the original sentence means. One believes propositions not facts (though one may believe in facts): propositions, not facts, are the proper objects of belief. But in the case of knowledge this is inverted: one doesn’t know propositions (except perhaps in the sense of being acquainted with them), but one does know facts. Thus we may conclude that belief and knowledge have different objects—propositions and facts, respectively. One believes propositions and one knows facts, and facts and propositions are different kinds of thing (we needn’t here go into exactly what kinds of thing they are).

            Then we can formulate a different theory of knowledge—call it the “factive theory”.  [1] Knowledge is a relation to facts not propositions, while belief is a relation to propositions not facts. Knowledge is not a propositional attitude but a factive attitude; hence it is not a form or type of belief. This doesn’t mean that belief is not a necessary condition for knowledge; it just means that knowledge itself is not logically a type of belief—as it might be, meritorious belief. To know something is not to believe a proposition of a special type (a true and justified proposition), but to stand in a relation to another type of entity altogether, viz. a fact. The knowledge relation is a sui generis relation not a special case of the belief relation. There are propositions and there are facts: belief maps onto the former, while knowledge maps onto the latter. The believing relation relates the subject to a proposition, while the knowing relation relates the subject to a fact.

            It is the same with perception and memory: these too are better handled by the factive theory than the doxastic theory. You can give “John perceives the fact that p” but not “John perceives the proposition that p” as a paraphrase of “John perceives that p” (if John perceives propositions at all, that is not the meaning of “John perceives that p”). And you can paraphrase “John remembers that p” by “John remembers the fact that p” but not by “John remembers the proposition that p”. So perceiving and remembering are not type of belief either, though there may be suitable beliefs in the offing. If I remember the fact that I went shopping yesterday, I presumably also believe the proposition that I went shopping yesterday: but these are not the same thing. The verbs “remember” and “believe” have a different grammar, a different logic. If we came to the view that perception, memory, and knowledge do not require an underlying belief after all—being psychologically more primitive than belief—that would not surprise the factive theorist, since he never thought they were special types of belief to begin with. There may not be beliefs and propositions in the picture at all, just facts and knowledge of them. Animals incapable of belief may still be able to stand in the relation of knowledge to facts–as when an animal knows it is about to be attacked but has no beliefs about the matter (it doesn’t form an opinion about its parlous condition).

            And then there is the matter of referential transparency and opacity. Suppose the sentence  “John knows the fact that Hesperus is a planet” is true: is “John knows the fact that Phosphorus is a planet” also true? I submit that the second sentence will also be true: John knows the fact underlying this kind of re-description (renaming). It is the same with “John perceives the fact that Hesperus is twinkling”: we can substitute “Phosphorus” here too. But notoriously we cannot make this kind of substitution inside belief contexts. We preserve the same fact under such substitutions, but not the same proposition. Therefore knowledge is not a type of belief: knowledge is referentially transparent, but belief is not. The context generated by “the fact that” does not have the logical properties of the context generated by “the proposition that”—because, intuitively, facts concern objects and properties in the world while propositions concern modes of presentation of objects and properties in the world. Mode of presentation (representation) doesn’t matter to facts, but it does matter to propositions—because propositions exist at the level of representation not at the level of things represented (sense not reference). Ascriptions of knowledge relate individuals to objective mind-independent facts, while ascriptions of belief relate individuals to representational entities. Hence the former ascriptions are transparent and the latter opaque.

            The sentence types “x knows that p” and “x believes that p” are superficially similar in form, which may fuel the doxastic conception of knowledge, but they differ in logical form and in the ontology presupposed. One expresses a relation between an individual and a fact, as revealed in the paraphrase containing “the fact that”; while the other expresses a relation between an individual and a proposition, as revealed in the paraphrase containing “the proposition that”. Different relations, different objects: so the former is not a special case of the latter.

Colin McGinn

  [1] I am using “factive” in a new sense—not the familiar idea that “knows” is a factive verb, i.e. implies the truth of what is known, but the idea that knowledge is to be understood as a relation to facts as part of its analysis.

                                                Internal Behaviorism

Goethe’s famous dictum, “In the beginning was the deed” is often invoked as a motto for behaviorism: if you want to understand the mind, you should look to overt behavior not to some supposed inner landscape. But that is not the only way the dictum can be interpreted, since not all deeds involve motions of the body—there are also mental deeds. There are intentional mental acts, such as remembering, imagining, and judging; these might be described as “mental behavior”—what the subject is doing mentally. This suggests the possibility of a distinctive new position in the metaphysics of mind—what I am calling “internal behaviorism”. It is not that I think this position is preferable to all others, or even that it is particularly plausible, but it is worth identifying it and exploring its contours. Think of this as a “research program”. So let’s examine its possible motivations and its strengths and weaknesses.

            The idea is that mental actions are ontologically primary in the analysis of mind. We have mental states, mental traits, mental faculties, and mental systems, but all these are to be understood in terms of the mental actions associated with them. Following classic philosophical behaviorism, we could formulate the doctrine by saying that these mental entities are to be analyzed as dispositions to produce mental actions. For instance, beliefs are dispositions to make judgments, memories are dispositions to remember, and personality traits are dispositions to choose. We need not assume that all mental behavior is fully intentional: we can allow for sub-intentional mental acts or even reflexive mental acts—just as with bodily behavior broadly construed (“responses”). We could even extend the notion of behavior to include mental events of all kinds: then the doctrine will be that all mental notions are to be analyzed by reference to mental events. We could accordingly maintain that an emotional state like fear or anger is a disposition to experience suitable affective events such as feelings of fear or anger. Everything mental is bound up with mental events, many of them actions, and the mind is best understood from that basis. States, traits, faculties, and systems are all entities whose nature is constituted by the kinds of mental actions or events associated with them—by the mental deeds to which they are tied. What the mind does is key to what the mind is.

            This kind of behaviorism might be part of a general metaphysical preference for events over other entities (substances, objects, properties) or it might be a doctrine specifically about the mind, holding that the mental is distinctively constituted by mental activity. Descartes viewed the mind as a mental substance imbued with mental attributes; on the behaviorist view (external or internal) it is better seen as a locus of action—a collection of animated events. Hitherto this metaphysical picture has been expressed in terms of bodily action, but that is not the only way to go: we can hold that the mind is essentially a site of mental activity—engaged not passive, flowing not fixed. There can be no mind where there is no mental behavior. What is an intention if not a disposition to form specific intentions to act? If I intend to become a good tennis player, I will form innumerable specific intentions to practice, study, and compete; if I didn’t I could not be said to have that intention. What reality could there be to the mind that is not manifested in the various activities of the mind? Aren’t mental actions and events the only things in the mind that can be introspected? When I know what I believe don’t I just know the episodic thoughts I am disposed to have? I can’t directly introspect my mental states, only the events that are associated with them. How can I know anything about my memories except by reference to the acts of remembering that manifest them? Introspection is knowledge of what my mind is doing.

            Internal behaviorism respects the active nature of mind while avoiding the problems afflicting external behaviorism. External behavior is neither necessary nor sufficient for having a mind (paralysis and robots), but surely mental behavior is necessary and sufficient for having the attributes of mind. How could we disconnect beliefs from dispositions to judge, emotions from dispositions to feel, or personality traits from dispositions to choose? To be mentally paralyzed is to lose one’s mind, and there cannot be a mindless robot endowed with mental activity (but no mental states, traits, faculties, or systems). We are analyzing one part of the mind by reference to another part of the mind, not trying to analyze the mind by moving outside the mental realm altogether. The mind does not reduce to muscular contractions, to be sure, but it might reduce to its own actions—as it were, mental contractions. Of course, internal behaviorism does not make the mind public in the way external behaviorism does; but that was a misguided and quixotic project. Indeed, we might well favor internal behaviorism precisely because it keeps the mind where it should be—resolutely private. The mind consists of inner behavior—hidden, directly known only to the subject, removed from the public world of bodies. The deed is basic, but it is a private deed.

            Once we have formulated this kind of inward-looking behaviorism we might venture into more sophisticated versions of the basic idea. Thus we might replace the idea of simple dispositions to mental behavior with something more like functionalism: a mental state is a functional state existing within a web of other mental states and connecting with mental actions and events. If I desire some chocolate and believe there is chocolate in the fridge, I will make the decision to go to the fridge; the belief and desire interlock and lead to the mental action of decision-making. Desires also function holistically in the production of choices, decisions, and intentions, so we will want to build that into our account of how desires relate to mental actions and events. The dispositions to mental behavior will be complex and interdependent; but analytic bedrock will consist of appropriate activities of mind. Thus the spirit of functionalism is maintained while its commitment to bodily behavior is dropped. The mind is a functional system, but the functions are defined over mental inputs and outputs—things like events of perception and actions of judging. We will have no problems of “absent qualia” or “inverted qualia” under such a conception, since the mental is part of the very analysis of the mental.  [1]

            There is really nothing oxymoronic about the phrase “internal behaviorism”, just certain dispensable prejudices. Mental acts are as real as physical acts—the realm of action extends into the mind: calculating, talking to oneself, imagining, problem solving, and so on. So there is nothing metaphysically problematic about trying to use mental actions to give a general account of the mind—reducing it all to mental activity. The view may be misguided, but it is not in principle ruled out by the very definition of “behavior”. No doubt classic behaviorists were motivated by anti-mentalist assumptions, but the core of the position is detachable from all that, namely that the mind is best understood in terms of activity. The mental deed is basic—mental doing precedes mental being. We should therefore add internal behaviorism to the list of logically available metaphysical theories of the mind.

Colin McGinn                 

  [1] We can either accept reductive internal behaviorism or eliminative internal behaviorism: we analyze everything mental in terms of dispositions to mental behavior or we simply deny that that there is anything to the mind except such behavior. The view suits an idealist attracted by the notion that mental being is a matter of deeds of some sort.

Contradiction

Contradictions arise quite naturally in several areas of thought and it is not easy to resolve them.  [1] It is generally felt that contradictions are unacceptable for two reasons: (a) contradictions entail anything and (b) they violate the logical law of non-contradiction. But some have questioned these reasons for rejecting contradictions, developing what is called “para-consistent logic”. There are even those who stoutly maintain that contradictions can be true (as opposed to necessarily false). For example, in the case of the Liar paradox a sentence can be both true and false (not true); and an author might describe a fictional character in contradictory ways (knowingly or unknowingly) both of which must be accepted as true. The point I want to make is that even accepting that kind of view we can still insist on the universal applicability of the logical law of non-contradiction, properly understood.

            Suppose we allow that a proposition of the form “x is F and x is not F” is true. We are then saying that an object x has both the property of being F and the property of being not-F, so that these two properties don’t logically exclude each other (of course, traditional logicians deny this, holding that these properties do exclude each other). What we are clearly not saying is that x lacks one of these properties: we are saying that x has both of them. Suppose now that we say explicitly that x lacks one of the properties, say the property of being F: “x is F and xlacks the property of being F”. That surely is impossible: no object can both have a property and lack it—if the property is present, it is necessarily not absent. We could call this law “the Law of Non-Exclusion”: if an object has a property, that fact excludes lacking that property. I don’t think there are any areas in which we are forced, or even tempted, to accept any exception to that law. In the case of the Liar paradox, accepting the contradiction as true is a matter of accepting that a proposition can be both true and false—it is not a matter of accepting that a proposition can be true and yet lacks the property of being true (similarly for the property of being false or not true). In other words, saying that an object is not F is not the same as saying that it lacks the property of being F, since the latter logically excludes being F while the former does not (if we accept true contradictions).

            Suppose an author describes a character, Ned, as tall early in her novel but as not tall later in the novel, having forgotten her earlier description. We then seem entitled to say that Ned is both tall and not tall—at least one can sympathize with the temptation to say that. But this is not tantamount to saying that Ned both has and lacks the property of being tall: he has both, though they contradict each other. It would be quite wrong to say that Ned lacks the property of being tall because he is later described as not tall, since the author’s earlier description is sufficient to confer tallness on Ned (like her later description of Ned as not tall). That is, accepting such contradictions as true, as some recommend, is not the same as accepting that objects can have properties but also lack them: it is accepting that objects can actually have contradictory properties. If this is right, then accepting contradictions as true is not in violation of the correct formulation of the logical law of non-contradiction, since that law states that objects cannot lack properties that they have. To accept that objects can have contradictory properties is not the same as accepting that they can have properties and also lack them; on the contrary, it is accepting precisely that they have both properties. So we can accept contradictions without violating the law of non-contradiction, correctly understood, i.e. the law of non-exclusion. Or perhaps we should say that there are really two laws of non-contradiction, one of which has exceptions (if we choose to go that way) and one that does not. To accept that there can be true contradictory statements is not to accept that there can be states of affairs in which objects both have and lack the same properties.

We thus have some wiggle room when it comes to areas that easily generate contradictions: it becomes easier to accept the existence of true contradictions—for example, accepting that the Liar sentence is both true and false. This does not require us to give up on the law of non-exclusion, which is arguably what the law of non-contradiction was saying all along. The disturbance is nowhere near as great as would be occasioned by accepting that something can both have and lack a given property. However, there is no reason to contemplate that possibility stemming from any of the areas that generate contradictions, and it is surely logically ruled out. We are naturally led to entertain propositions of the form “x is F and x is not F”, but nothing requires us to contemplate the (alleged) possibility expressed by “x is F and x lacks the property of being F”. In the problematic cases these two propositions come apart.

            Contradictions don’t arise everywhere but only in special cases; maybe they could not arise everywhere. Contradictions are restricted in scope. This means that we need not tinker with the law of non-contradiction for the majority of cases, since there cannot be naturally occurring contradictions in these cases. It is only in certain special cases that we might contemplate accepting true contradictions: but even here we need not contemplate abandoning a universal law of logic, namely the Law of Non-Exclusion. We can tolerate contradictions without tolerating objects simultaneously having and lacking properties.  So the para-consistent logician need not be quiteas radical as he appears to be.  

  [1] I won’t discuss any of these areas in detail, but I have in mind not only the semantic and set-theoretic paradoxes but also the following: contradictions arising in fiction, contradictions arising through vague concepts, and Kripke’s argument that our practices of belief ascription license contradictory ascriptions of belief. In all these cases there are apparently cogent reasons to endorse contradictions—reasons to accept contradictions as true. My question is how drastically revisionary it would be to succumb to those reasons.

Difference and Necessity

Essentialists have found in identity a shining example of their creed. Any object is necessarily identical to itself—how could that not be so? It has even been thought that the necessity of identity can be proved from Leibniz’ Law: given that a is necessarily identical to a and that a is identical to b, it follows that b is also necessarily identical to a. But far less attention has been paid to the modality of difference: if a and b are different objects, is that a necessary fact or a contingent fact? If objects are distinct in the actual world, are they distinct in every possible world? Given that I am actually distinct from Saul Kripke, are there any worlds in which we are identical? Is there any world at which the statement “Colin McGinn is identical to Saul Kripke” is true? We can clearly lack certain of our actual properties in a possible world—neither of us are philosophers in some possible world–but could we lack the property of being numerically distinct? Are there worlds in which London is identical to Paris or the moon is identical to the sun? Could twins be numerically identical in a possible world?   

            The answer is intuitively obvious: none of these things is possible. If objects are different, they are necessarily different—different in all worlds in which they exist. If a is different from b, this is part of a’s essence—it is essentially different from b. Thus objects have an enormous number of difference essences, as many as there are objects from which they are distinct. They only have one identity essence, viz. identity to themselves, but they have hugely many difference essences, according to all the objects in the universe. In fact, they have infinitely many difference essences, given the infinity of numbers or points in space: I am different from the number two, say, in every possible world, and so for any number. These essences come cheap and plentiful. Furthermore, they are a distinct species of essence, not to be assimilated to the other kinds—origin, composition, etc. Nor are they derivable from the necessity of identity; they cannot be proved from Leibniz’ Law, for one thing. The difference of a given thing from everything else is essential to what it is, as a primitive modal fact. No amount of merging or fusion can make two actually distinct things into the same thing. For instance, you might think two rivers that flow side by side could be one river in a world in which the dry land separating them is removed; but such a world is really one in which a new river exists, not a world that contains the old two rivers joined in a single river. Those actual rivers don’t exist in this possible world; we have fused two rivers into a broader river and eliminated the actual rivers. Physical merging or joining is not two things existing as one, for those things will no longer exist—any more than fission is one object existing as two, and hence not being self-identical. There are no worlds in which McGinn and Kripke both exist and yet are one person—though there is a world in which an individual has characteristics belonging to each of us.

            That is all at the level of metaphysical possibility—what could really be. I have said nothing yet about epistemic possibility, and here matters assume a different shape. For it is an epistemic possibility that McGinn and Kripke are the same individual: it might turn out that we are identical—some very elaborate trick has been performed. To take a simpler example, suppose I am dating a woman I believe to be an only child but in fact she has an identical twin. I have been seeing both twins but don’t realize it: I have mistakenly believed of different women that they are one woman. Isn’t it possible that this is true of a great many people I know? I am always assuming identity from one meeting to the next, but in fact there is a plurality—different people where I thought there was one person. We can conjure ample grounds for a skeptical argument here. So it is not an epistemic necessity that what I take to be one individual is really one individual—it might turn out that I know many more people than I realize. Just as you can mistakenly suppose that what is in fact a single object is two, as in a classic Frege case, so you can mistakenly suppose that what are in fact two objects are a single object. Judgments of identity and difference are fallible. But none of this has anything to do with questions of metaphysical necessity: just because one object can be mistaken for two, or two objects mistaken for one, has no bearing on whether identity and difference are metaphysically necessary relations. We should not confuse metaphysical and epistemic questions. How things might turn out is a quite different question from how things could be. It might turn out that there are many things where you thought there was one, or that there is one thing where you thought there were many, but these possibilities don’t show that one thing could be many things or many things one thing.

            Frege used identity statements to prove that distinct senses can correspond to the same reference, since this is what explains someone’s inability to know the identity a priori. But an exactly parallel argument leads to the conclusion that there can be many references corresponding to a single sense, namely that a person can think he is referring to one object when he is really referring to two. If I call the woman I am dating “Joan” in the morning and “Joanie” in the evening (she always seems more fun-loving in the evening), I will subscribe to the truth of “Joan is Joanie”, even regarding it as analytic. But given that I am seeing twins and don’t know it, we have to say that I really have two references, while meaning the same by the name. The twins are perceptually presented to me in the same way and I intentionally use the names “Joan” and “Joanie” as synonyms, but I refer to different individuals when I use the name in relation to the women in question. This is a case of sameness of sense combined with difference of reference. To me the statement “Joan is Joanie” is analytic, but the context of use generates distinct references.

            The same points apply to natural kinds: people can fail to realize identities and fail to realize differences. You can fail to realize that water is identical to H2O and you can fail to realize that gold is different from copper (you think they are the same metal). Nevertheless, it is part of the essence of water to be H2O and part of the essence of gold to be different from copper. There are no worlds where water is identical to some other chemical compound and no worlds where gold and copper are the same metal (though one metal might be called both “gold” and “copper” in a possible world). If pain is identical to C-fiber firing in this world, then it must be so in all worlds; and if sensations of red and sensations of green are different in this world, then they must be different in all worlds. In order to create pain all God had to do was create C-fiber firing (according to the identity theory); and in order to create gold and copper God must perform two acts of creation—he can’t skip one of these creative acts and more economically create both metals in the form of a single metal. Different things can never be other than different, and identical things can never be other than identical. Hesperus is Phosphorus in every world and Venus is not Mars in every world—despite the fact that both these propositions are not known a priori. Modality behaves in the same way for both identity and difference: the essentialist can also find in difference a shining example of his creed. We find necessity not just in the relation a thing has to itself and to no other thing but also in the relation a thing has to other things and not to itself.