A List

I will make a list so that you can see that I’m not exaggerating.

Intellectual. Philosophy: all areas, technical and popular. Science: psychology, biology, economics, physics. Literature: Shakespeare, Jane Austen, Nabokov, etc. Writing: novels, short stories, poetry, songs (rock, ballads, blues, pop).

Sports. Tennis, table tennis, squash, badminton, gymnastics, pole vault, discus, basketball, football, cricket, archery, knife throwing, swimming, kayaking, surfing, windsurfing, kiteboarding, skim boarding, paddleboarding, skiing, ice skating, bowling, darts, skateboarding, mountain biking, motorcycling, trampoline, body building.

Music. Drums (rock, jazz, djembe), guitar, harmonica, voice (rock, ballads, blues, pop).

Epistemic Necessity and the Good

The connection between metaphysical necessity and the Good is speculative and questionable, though there are signs of affinity.[1] But the connection between epistemic necessity and the Good is immediate and easy to discern: it goes via certainty. If a given proposition is epistemically necessary, it is certain: I am certain that I exist and that I think. To be epistemically necessary is to be certain. By contrast, epistemic contingency is the same as doubtfulness: you might turn out to be wrong. It is the difference between confidence and diffidence, being sure and being unsure. Certainty is good and uncertainty is not so good (sometimes quite bad). We would all like to be (justifiably) certain of everything, if only we could; but we live with uncertainty, however uncomfortably, there being no alternative. Uncertainty is a necessary evil. We therefore seek certainty and try to avoid uncertainty: we prize epistemic necessity. We like the idea that we can’t be wrong—that our reasons necessitate our conclusions. We value the necessity of logical entailment. This is what gives rise to epistemic necessity. If nothing ever entailed anything, there would be no such thing as certainty. Logic is a good thing.

But why is certainty good? Two ideas come to mind: certainty is instrumentally good in relation to desire; and certainty feels good in itself. If we can’t be wrong, we can’t be wrong about what will satisfy our desires; in the ordinary sense, I am certain the supermarket is open on a Sunday, so I go there to pick up food—and I am never disappointed. But if I am rationally doubtful, I might end up hungry. This is the basis of epistemological pragmatism—pragmatism about the value of certainty. Alternatively, it might be maintained that it feels good to be certain and not good to be uncertain—the former is more pleasurable. I think this is true and not a negligible point, but I don’t think it is the end of the story. It isn’t simply a matter of hedonism—the higher pleasures and all that. Something more interesting is going on, which gets to the heart of epistemology. We value certainty because we value knowledge, and knowledge requires certainty. What kind of certainty is a debatable point—philosophical certainty or commonsense certainty. In the ordinary sense, we have a good deal of certainty, as in the supermarket example, so we have a good deal of knowledge. That’s good, because knowledge is good. Epistemic necessity gives us knowledge, and we happily accept the gift. Doubt undermines knowledge, and the more doubtful the more the undermining. But why is knowledge good? That is a very good question. Is knowledge to be defined as something like true justified belief or is it more like direct perception of a fact? The former leads naturally to a pragmatic conception of the value of knowledge—knowledge is what gets you fed (so banal, so boring!). The latter opens up a more interesting proposal: knowledge is good because it connects the self to the world. We value knowledge because it is direct apprehension of reality—and we value this in its own right not because of pragmatic consequences. We value knowledge for the same kind of reason we value friendship or romantic love—because it relieves existential isolation. So, we value epistemic necessity because we value certainty, and we value certainty because we value knowledge, and we value knowledge because we value connection. We like to merge (“Only connect!”). And we do merge: perceptual consciousness presents the objective world to us as separate but also possessed. According to the Cogito, I am connected in this way to myself—not remotely but immediately. It would be terrible to be disconnected from myself, as if I were a mere hypothesis (like black holes or dark matter or fairies). It would also be good to be at one with my environment—and I am by virtue of seeing it and touching it. I am not cut off from it, blind to it, ignorant of it: I know it, as I know myself. That is, I enjoy acquaintance with my environment, as I enjoy acquaintance with other people and animals. I am not stuck inside my own inner world, just a passing show of inchoate sensation—that would be a kind of hell. This kind of perceptual acquaintance is part of what makes life valuable, worth living. Luckily, we have quite a bit of it (pace the philosophical skeptic). We are not blindly guessing from a distance, hoping against hope that there is something out there. It’s good to be in the know, well acquainted, in with the in crowd, part of life’s rich pageant—not a stay-at-home, a shut-in, a reluctant lonesome cowboy. Knowing is part of human existence, perhaps the main part. But not that etiolated sort of knowing beloved of the analytical philosopher (true-justified-belief knowing) but embodied-direct-perception-of-actual-concrete-facts knowing, as in seeing the sun rise in the morning.[2] We have epistemic connection to thank for that. It is something quite profoundly meaningful. I am at one with the world not a self-enclosed particle. Perceptual knowledge has human value.

Does all this have any bearing on metaphysical necessity? None at all, you might reply, these being quite different concepts. You are not wrong there, to be sure, but is there no connection at all between the two concepts? After all, the word “necessary” is used for both (which has caused a lot of confusion about the distinction); and the epistemic use might spill over into the metaphysical use by association. But more forcefully, epistemic necessity is defined by metaphysical necessity (but not vice versa): for it involves the idea of a set of reasons entailing a conclusion, and entailment is a metaphysically modal notion, i.e., the reasons necessitate the conclusion. In all possible worlds in which the reasons obtain the conclusion obtains.[3] There is such a thing as epistemic necessity only because there is such a thing as metaphysical necessity. We therefore can’t value the former without valuing the latter. There is only knowledge because metaphysical necessity exists, i.e., premises entail conclusions. Certainty requires that the grounds necessitate the inference, as in the Cogito. To put it differently, there has to be a type of non-epistemic necessity for there to be epistemic necessity (even if it doesn’t add up to what the logicians regard as necessity). You therefore can’t value the one without valuing the other. The case is similar to nomological necessity: there can’t be epistemic necessity (certainty) without nomological necessity, since we rely on laws of nature in all our reasonings, so we can’t value epistemic necessity without valuing nomological necessity. It’s good to be certain that the supermarket is open on Sunday, but this requires that there are laws of nature concerning shop openings, worker availability, motion, etc. Nomological necessity is good because it enables epistemic necessity (inter alia), which is itself good. We are glad there are laws of nature otherwise we couldn’t be sure of anything, and we like to be sure. We like to be sure because we like to know. And we like to know because otherwise it would be a pretty dismal isolated existence. Plus, it feels nice to know things and it gets you fed. There are reasons why necessity is regarded as good.[4]

[1] See my papers “Is Necessity Good?” and “Necessity and Change”.

[2] See my “Perceptual Knowledge” and “Non-Perceptual Knowledge”.

[3] See Kripke’s discussion of epistemic necessity in Naming and Necessity.

[4] It is also true that I know what I know necessarily in central cases. For example, if I am seeing a certain color and I know that color “by acquaintance”, I necessarily know it—I can’t help knowing it (compare pain). Necessarily, if I am acquainted with X, I know X. Likewise, I necessarily know that I exist, by the Cogito: I can’t not know it. This is another way in which metaphysical necessity is connected to epistemic necessity. In general, if I am certain that p, I must know that p. If I’m certain that I’m thinking, I can’t not know that I’m thinking—the knowledge is forced on me. There are many metaphysical necessities of epistemic truths, i.e., things I can’t help knowing.

A Story About John Searle

In 1984 John Searle gave the Reith lectures in England. This was a big deal and quite an honor. I was asked by the BBC to join a small panel commenting on these lectures. After the program was recorded, we went to a boozy lunch set up by the Beeb. It was, shall we say, pretty lively. It culminated in John standing on his chair to make his point more emphatically. This is the kind of behavior I approve of (I was with George Soros once in St Barts at a restaurant in which we all stood on the chairs and table at the end of the night to dance, him included). How things have changed. I also had a brief correspondence with Sir Peter Strawson on the best use of the word “fuck” in literature (I cited a passage from Martin Amis). Those were the good old days. He also said to me once at an academic gathering in Oxford, “There are people here I wouldn’t mind talking to, but none I want to talk to”. He also said of Putnam, then visiting Oxford, “Hilary Putnam is a very sweet man, as he would be the first to agree”.

Attention and Knowledge

In chapter XI (“Attention”) of The Principles of Psychology William James makes the point that the empiricists completely omit the contribution of attention in the generation knowledge. The reason is that attention possesses “a degree of reactive spontaneity” that breaks through the “circle of pure receptivity which constitutes ‘experience’” (402). Instead, he says, “My experience is what I agree to attend to” (his italics). Evidently, this is a good point: we only form knowledge in the normal course of events if we attend selectively to the object known about. We don’t acquire knowledge of things to which we pay no attention. Attending is a necessary condition of knowing. But attention doesn’t come from the world but from the knowing subject. It isn’t an “idea” deriving from an “impression”, still less a facet of the external object: it isn’t a stimulus. It is endogenous not exogenous. Presumably it is innate: the organism is innately primed to attend to some things and not others—the things that matter to its survival. Consciousness focalizes, concentrates, highlights. It doesn’t just passively reflect whatever impinges on the organism’s senses, indiscriminately. No attention, no knowledge, or very little. Attention is what provides cognitive uptake. A better version of empiricism would assert that all human knowledge derives from attention to the data of sense, and thus depends on innate features of the organism. It is therefore not a pure empiricism: the mind is not a blank slate but a self-generated spotlight focused on the objects of sense. Knowledge depends on the mind actively reaching out not just passively receiving imprints from outside. The theory might better be called “perceptual attentionalism”.[1]

The point carries over to the analysis of knowledge, as distinct from its origins. If to know is to perceive[2], then we must add that the perceiving must be attentive perceiving not merely peripheral or unconscious perceiving. That is, the primary knowledge we have is attentional perceptual knowledge; anything else is secondary. The paradigm of knowledge is attending selectively to a sensory object. Not true justified belief in a proposition, whatever that amounts to, but attention to a perceived stimulus—then we really know (“acquaintance”). This is the foundation of all genuine knowledge, the sine qua non. We are not seeing or hearing in an unfocused or distracted manner, but are focused and concentrated—drinking it all in, as we say. To that degree knowledge is a voluntary act: the act of knowing is the act of attending. You can try to know the object better by focusing on it longer or harder. Epistemology must not neglect this aspect of knowledge by focusing (!) on belief and justification. Seeing and attending are the basic epistemic operations. Knowing creatures can do these things even when belief and justification are beyond them. Strictly, then, we shouldn’t speak of the perceptual theory of knowledge but of the perceptual-attentive theory. Knowledge is the upshot of sensation and attention working together.

How does rationalism deal with attention? It is as silent as classical empiricism (James says nothing about rationalist epistemology). Tacitly, however, it helps itself to the faculty of attention: for attention is how innate information (for want of a better word) becomes conscious knowledge (as in Plato’s Meno). The learner attends to what is innate and thereby converts it to conscious knowledge. So, the rationalist theory of the origin of knowledge is really a combination of innate “ideas” and acts of inward attention. Attention is necessary in order that rational a priori knowledge should exist. Without attention the innate endowment would remain unconscious, merely latent. The acquisition of mathematical knowledge, for example, depends on the ability to attend selectively—to focus. It doesn’t arise from the innate material alone. Both a priori and a posteriori knowledge therefore require an active attentional faculty, not mere “experience” and “innate ideas”. The classical theories need to be enriched with this element. Attention is the medium (midwife) of knowledge; experience and innate information provide only the raw material. Without attention we would know nothing worth knowing. It may well be that our superior state of knowledge, compared to other animals, is the result of greater powers of concentration, i.e., selective attention. Attention is what actively and essentially mediates between the senses and knowledge proper. This is why the teacher exhorts his students to “Pay attention!”, because without attention knowledge cannot be formed. And we all know that what we don’t attend to doesn’t make it into memory. It is true that the passive imprint model is hard to discard, but we need a more active internalist conception of knowledge—its origins and nature. Knowledge doesn’t just happen to us; we make it happen.

But attention is quite puzzling. What is it exactly? We can study it experimentally, we can examine its neural correlates and behavioral expression, but we have little to say about it directly beyond metaphors (searchlight, spotlight). It is a touch mysterious, is it not? This is not surprising given its close connection with consciousness. Perhaps that is why theorists have been reluctant to invoke it in their theories. But nobody doubts its existence or importance, so it might profitably figure in our theories. This seems to be the situation: attention is integral to knowledge, but we don’t know much about it.[3]

[1] No one has ever explicitly espoused this theory to my knowledge.

[2] See my “Perceptual Knowledge”.

[3] Could we define attention as “the faculty of knowledge acquisition”? Is that its evolutionary function? It evolved to facilitate the retention of information. What came first is not belief or even knowledge but attention—the means by which knowledge is acquired. Organisms went from unfocused primitive awareness to focused cognitive pick-up. This is the right evolutionary epistemology. It may also characterize the development of knowledge in the child’s mind. The ability to attend is what defines what we call intelligence. We could establish a new school: attention-based epistemology. Educationally, we need to find ways to encourage and improve pupil attentiveness. It would therefore be nice to know more about it (“attention science”).