You call yourself a disenchanted naturalist and me an optimistic naturalist; I would like to propose a slight adjustment to your contrast. I am just as disenchanted as you are—I don’t think that any of the features of our dear manifest image, the world we live in, is enchanted, imbued somehow with a special well-nigh magical property or power that defies explanation in terms of atoms and molecules and the laws of physics, but I don’t think this recognition calls for any pessimism.

I am not in the slightest disappointed that the accounts we can give of the manifest image boil everything down, in the end, to the laws of physics, to efficient causation. All hail glorious reduction! The cherished peculiarities of the manifest world are quite wonderful enough without being magical; indeed, the more I learn about how Darwinian processes must have generated the manifest image, the more wonderful it seems.

Consider a happy quartet of adjectives: sweet, cute, sexy and funny. These quite obviously do not attribute properties to atoms and molecules, on a par with mass and charge. Sweetness, we now recognize, does not reside as a mysterious “intrinsic” property of glucose, but as a functional property established by evolution to encourage us animals to seek out high-intensity energy to fuel our lives, and the cuteness of babies is a similarly useful provoker of solicitude and protection for an altricial species. Sexiness goes without saying, and why anything at all should be funny is a riddle I think Matthew Hurley, Reginald Adams and I have now solved (in Inside Jokes: Using Humor to Reverse Engineer the Mind, MIT, 2011). Yes, Darwin’s great idea of natural selection is universal acid, and among the features of the world it dissolves entirely are these “intrinsic” properties of macroscopic things. But that doesn’t leave the world bereft of sweetness, cuteness, sexiness and humor. It accounts for why we enjoy these features of the world we are designed to live in. They are no more illusory than colors or solidity or weight.

Perhaps chief among the items reduced by Darwinian universal acid is purpose. You say that “Darwin substituted Aristotle’s efficient causation for the purposes that everyone since Aristotle had employed to explain the nature of living creatures. What we call biological functions turned out to be a small subset of a vast range of local adaptations, most of which we haven’t even noticed yet. Darwin showed that the clockwork universe extended beyond the physical domain to the biological one.” True enough, but you go on to say “There are no purposes in nature.” No Aristotelian purposes, I grant you, but why take Aristotle’s ancient vision as the anchor, today, of the concept of purpose? You offer a vivid analogy:

. . . treating blind variation + environmental filtration as purpose is just Orwellian Newspeak. It would be like saying that Kelvin explained caloric as molecular motion or Einstein vindicated Newton’s occult force, gravity, by showing it’s just the curvature of space. No. Kelvin explained away caloric, and Einstein explained away gravity. Mutatis mutandis, Darwin explained away purpose.
But there is flaw in your analogy, clearest in the case of Kelvin: he did explain away caloric, but he didn’t explain away heat. We could argue over whether Einstein explained gravity in terms of curvature of space or explained away gravity in terms of curvature of space. And I could—and will—argue that Darwin explained the origin of purpose. This is not “Orwellian Newspeak.” It is a straightforward account of a phenomenon that can be seen from the perspective of cosmology, not just from some localized, subjective perspective of the manifest image.

David Deutsch has recently spoken eloquently about this in a TED talk about what he calls the Great Monotony: the billions upon billions of years after the Big Bang where nothing new happened. And then things picked up dramatically. The pivot point was the beginning of natural selection, which led to ever more capable structures, made of the same atoms and molecules that had been shuffling around for eons, but now able to form structures that could reproduce themselves and compete for resources and refine their designs and discover ways of extracting information from the world around them to utilize in still more capable designs. And the puzzle that will never be answered by mere physics is how some of these structures were capable of inventing physics. The reflexivity of inquiry is a very recent addition to the furniture of the universe. It is a product of purposeful exploration and testing that has blossomed within the last few millennia, after billions of years of utter clueless stupidity, and a few billion years of myopic life on our planet.

The merely Darwinian creatures, such as bacteria, have led to the Skinnerian creatures, such as worms, and then to Popperian creatures, such as birds and mammals (we can argue endlessly about whether various apparent signs of insight in other creatures warrant elevating them to the ranks of the Popperians), but finally we get to us Gregorian creatures who benefit from the dual pathways of genes and culture, equipping our explorations with tools we didn’t have to invent, facts we didn’t have to discover from our own experience, and able to compare notes critically with all our conspecifics.

Yes, our Gregorian powers are ripe for demystification, and that is in effect what my life’s work has attempted. But that doesn’t mean that we don’t have purposes. One of your purposes, misguided though it may be, is to persuade me to join you in your pessimistic perspective, and my purpose here is to highlight the happy vision of the manifest image that I enjoy. Try it, you’ll like it.

Daniel Dennett

Daniel Dennett

You mistake the object of my pessimism. I’m not actually downcast about the fact that we can’t reconcile the “dear manifest image, the world we live in,” with science. So, maybe ‘disenchanted’ is the wrong word. Maybe ‘disabused’ better describes my movement away from the anthropocentric world view that science has been unraveling since Galileo. I’m certainly not disappointed in the reduction to physics of the world most of us think we live in. But that’s mainly because as far as I can see, it hasn’t been reduced, can’t be reduced, and that’s the problem that generates its irreconcilability with science. If only things did boil down, you and I would have no argument.

I can’t argue about whether the peculiarities of the manifest world you cherish are wonderful or not. Wonder, and wonderfulness is in the mind’s eye of the beholder. Besides, I’m as besotted by sweetness, cuteness, sexiness and funniness as any mortal. The disenchanted or disabused naturalist relishes them as much as any dualist.

Darwin’s theory does indeed account for why we enjoy these features “of the world we are designed [your harmless metaphor] to live in.” You say they are no more illusory than colors or solidity or weight. It’s an instructive comparison. Philosophers have been arguing about the nature of color, solidity, and weight (not mass) at least since Locke, just because Newtonian science couldn’t find them out there in the real world. They became secondary qualities, properties of the mind, qualia—the mystery mongers call them--that have awaited reduction since the 17th century. We both agree that Darwin’s theory accounts for why we enjoy feelings like funniness and sensations such as sweetness, indeed why they exist at all as spurs to adaptive behavior. But, alas, the theory doesn’t tell us exactly what sweetness, cuteness, sexiness, funniness or color, solidity and weight really are, in our brains. We’d like to get everyone to accept that they are functional properties, nothing but neuronal wiring patterns selected for their adaptative consequences. But despite our valiant efforts, wise-guy philosophers keep pointing out how the manifest image of these very feelings, their alleged qualia, make that reduction to Darwinian function controversial. If we could meet their objections, one large part of the manifest image would smoothly fit into the scientific one. If we can show that we don’t need qualia in our description of reality, so much the worse for the manifest image that is stuck with them. And if we can’t show this, so much the worse for naturalism, your kind or mine.

Let’s not argue about whether the manifest image and Aristotle shared the same ‘purpose’ or not. But let’s argue about whether natural selection is enough to give you what you really need to reconcile science and the manifest image. Because we both agree, that’s all we’ve got to work with. In an important essay you wrote that “Evolution by natural selection is a set of processes that ‘find’ and ‘track’ reasons for things to be arranged one way rather than another….natural selection starts with how come and arrives at what for…. It starts in “a lifeless world in which there are lots of causes but no reasons, no purposes at all…until at some point (but don’t look for a bright line) we find it appropriate to describe the reasons why somethings are arranged as they now are.” Actually you need a stronger conclusion: it’s not just that, from our stance, the reasons look appropriate (to us), but there’s got to be an objective normativity about them. The reasons have to provide objective justification for the adaptational outcomes Darwinian processes produce. It’s the natural emergence of these reasons (free free-floating rationales I think you called them in Darwin’s Dangerous Idea) that convert How come questions into What for explanatory demands.

(By the way, it’s that need for reasons to objectively make sense of outcomes—their telos--that anchors our sense of purpose to Aristotle’s.)

But I just can’t see how piling on more and more 2d law of thermodynamics-entropy-increasing processes, dissipating more and more order, to keep bigger and bigger molecular systems farther and farther away from chemical equilibrium, can convert causes into reasons. I can see how the process can make the design stance more and more useful for creatures like us. But that neither requires nor warrants treating the real patterns natural selection produces as the ones justified by the circumstances, which are therefore the reasons for them.

What’s so special about blind variation and environmental filtration that by itself can take us from How come to What for? You’ll say (have said) there is no bright line, “a proto-Darwinian algorithm morphing into a Darwinian algorithm…with no essential dividing line between them.” All we disenchanted, disabused naturalists see is the same old substrate-neutral universal Turing-machine (William Calvin called it a Darwin-machine) churning away.

Here finally we come to the essential disagreement between us, the toughest nut for us to crack. You concluded that essay:

“We can have our cake and eat it too. We can use the intentional stance [that the manifest image vouches safe to us] with a clear conscience, but only because Darwin has shown us how to cash out the intentional language in suitably austere talk about algorithmic processes of design generation and refinement.”
If Darwin (or his teleosemantic successors) had indeed shown how to cash out the intentional stance, it wouldn’t just be a stance. It would have all the reality of digestion or respiration or any other wholly chemical process. You yourself accept that “we use what for speculative hypotheses” generated by the intentional stance, “to help us frame testable how come hypotheses to test.” Those how come explanations (including the Darwinian ones) are the only ones science ultimately accepts.

So, this is what our disagreement really comes down to. Can we naturalize reasons—sentences, statements in our heads, that justify other sentences and statements. What your brand of naturalism needs to show is how reasons can be causes. You’ve got to show how the content of our beliefs can make a causal difference in the world. It’s not enough to show that the having of reasons--the brain states that having reasons consist in--bring about our behavior. What’s got to be shown is how come they justify it.

Alex Rosenberg

Alex Rosenberg

Your second letter clarifies your view for me nicely. It is the failure of reduction (in your eyes) that bothers you. We agree that if only all of the manifest image of everyday things could be placed uncontroversially into alignment with the scientific image, we could declare an important philosophical victory, but you see purpose, and/or meaning, as embarrassingly recalcitrant: we can’t do without them and we can’t show how they fit in with the rest of science. Quine notoriously described the impasse, under its familiar guise as Brentano’s thesis of the irreducibility of intentional idioms:

One may accept the Brentano thesis either as showing the indispensability of intentional idioms and the important of an autonomous science of intention, or as showing the baselessness of intentional idioms and the emptiness of a science of intention. My attitude, unlike Brentano’s, is the second. (19 60, p221)
And I took Quine’s peremptory dismissal with more than a few grains of salt. In fact, my whole career as a philosopher has been the attempt to describe a middle road between Brentano and Quine, as I described in “Mid-Term Examination: Compare and Contrast” the last chapter of The Intentional Stance (1987). I think I’ve succeeded, and you think I haven’t. Let’s see how you put it.

A key to understanding your current position is the point you make about qualia. As you say,

“despite our valiant efforts, the wise-guy philosophers keep pointing out how the manifest image of these very feelings, their alleged qualia, make that reduction of Darwinian function controversial . . . If we can show that we don’t need qualia in our description of reality, so much the worse for the manifest image that is stuck with them.”

In this you seem to be overlooking a real and attractive way of putting it: qualia are user-illusions, ways of being informed about things that matter to us in the world (our affordances) because of the way we and the environment we live in (microphysically) are. They are perfectly real illusions! They just aren’t what they seem to be; they are not intrinsic, unanalyzable properties of mental states; they are highly structured and complex activated neural networks that dispose us to do all sorts of things in response—such as declare that we’re seeing something blue.

The key move is to recognize that we have underprivileged access to the source or cause of our convictions about what we experience. You and I agree that we are endowed by natural selection with functional properties that populate our immediate world with relatively simple things with relatively simple properties (such as red chairs and blue sky and cute babies and painfully hot soup). The idea that our being thus informed about these familiar things is caused by our being somehow directly or immediately acquainted with “inner” mental properties is just a bad theory that glosses over our ignorance of how this is done. (Answering that question is one of Chalmer’s ‘easy problems’ but it dissolves his “Hard Problem”) So I don’t think qualia are any embarrassment to my naturalism.

What though of meaning, or, as you put it, reasons? I’ve given you an account of how natural selection gradually turns how come explanations into what for explanations (the full account is in From Bacteria to Bach and Back), but you see this as a subjective or parochial resting place for explanation:

“. . . it’s not just that, from our stance, the reasons look appropriate (to us), but there’s got to be an objective normativity about them. The reasons have to provide objective justification for the adaptational outcomes Darwinian processes produce.”

I agree, but I think they do! The designs of (the working parts of) organisms are just so good at what they are for! You grant that the Darwinian algorithm churns out “bigger and bigger molecular systems farther and farther away from chemical equilibrium,” but you don’t see how this “can convert reasons into causes.” The eagle’s eye is a measurably better seeing thing for an eagle than a sheep’s eye or a fish’s eye would be, and there are reasons for that. There can be no serious skepticism—can there?—about what an eagle’s eyes are for. As I have said, natural selection uncovered reasons long before there were reasoners to appreciate them.

Function is, one may say—and I have said—in the eye of the beholder, but the beholder may fail to see what is obvious. Here is where my allegiance to Quine’s embattled thesis of the indeterminacy of radical translation comes into play. Quine claimed that in principle, there could be rival translations of the utterances of one language into another and no hidden fact would settle which was right. This has struck most theorists of language as bizarre, but I think Quine was making an important point, which I have illustrated with a simple analogy. Consider a crossword puzzle that has two different solutions, and in which there is no fact that settles which is the “correct” solution. (The composer of the crossword went to great pains to devise a puzzle with two solutions.) It is in fact quite difficult to create a good example, though there are a few extant ones, including the simple four-by-four square one I concocted and published in my book Breaking the Spell (of all places). If making a simple crossword puzzle with two solutions is difficult, imagine how difficult it would be to take the whole corpus of human utterances in all languages and come up with a pair of equally good versions of Google Translate that disagreed!

Here we can ring in what I have called the cryptographer’s constraint: if you can find one reasonable decryption of a cipher-text, you’ve found the decryption. The odds against there being another good decryption or translation are astronomical. But that is the point; we can rely on this practical impossibility and not try to cement it in place with some imagined internal fact that settles the matter of the right meaning once and for all.

Every day we bet our lives on the hunch that we and our friends and neighbors mean roughly the same things by the words we use, and that the risk of an utter failure of radical translation is negligible. Doesn’t that give us meanings enough and reasons enough to carry on without attempting or hoping for a reduction of all this to microphysics? Our worlds of scientific explanation and mathematical proof depend on our reliance on this assumption. So far, so good. And thanks to the language we share we can turn the representation of reasons into causes with great reliability. If I say to you “Don’t open that box; it contains angry poisonous snakes that will leap out on you!” I have just presented you with a reason (articulated in language, of course) for not opening the box, and if you believe me, you will be caused by my utterance to avoid opening the box, which otherwise might have been a highly probable behavior. You wouldn’t be caused to refrain from this behavior if you didn’t trust me, believe me, understand me, and it is because we are such wily and skeptical interlocutors that our actions tend to be justified and not just caused by the reasons presented. Thus mere gigantic constructions of out-of-equilibrium quadrillions of molecules can be moved by reasons, as Kant would say.

Suppose I say loudly to a bear that is approaching me “Begone at once, or I will destroy you with my invisible light saber!” this may cause the bear to leave, but not because it is moved by the reasons I apparently offered it. That is how evolution, including the evolution of language, turns causes into reasons into causes. It would be foolish to try to explain this at the physico-chemical level—the physical stance—but the intentional stance makes this straightforward even to a child.

Daniel Dennett

Daniel Dennett

We are getting to the heart of the matter between us.

Purpose and meaning are, as you say, embarrassingly recalcitrant. But the embarrassment is not the naturalists’. These notions are just the last bastion of resistance to the scientific world view. Science can do without them, in fact it must do without them in its description of reality. It does so even as paleoanthropology explains with increasing detail how they and the rest of the manifest image emerged as convenient instruments of Pleistocene survival, but little more than that. Attributing purpose to ourselves, to our predators and our fellow scavengers solved a design problem. It got us from the bottom of the African food chain to the top in a matter of a couple of hundred thousand years. Then this quick and dirty solution got wired in and kept a grip on Homo sapiens till ultimately it dominated every aspect of our civilization, including its least and most barbaric institutions.

These notions have outlived their usefulness as instruments in the toolbox science employs to describe reality—including human psychology. But they will never loosen their hold on those who follow Brentano, the phenomenologist, as opposed to Quine, the scientific realist. And that’s most of us most of the time in the practical affairs of life.

It’s the play of images and silent word-sounds across subjective experience, endowed with the qualia we both disparage, that tricks the manifest image into its attachment to meaning and “conscious” purposes. The play of images and sounds, together with their efferent causes in external stimulation and their afferent effects in subsequent behavior seduces everyone’s manifest image into the illusion that thought, sensation, emotion all have content. But it’s all just a bluff that Quine recognized.

Ever since Descartes western philosophers have explicitly sought to vindicate introspection as a source of some kind of knowledge or other. They tried, for a couple of hundred years, without success to build the association of ideas in the head—images, sounds, feelings—into sentences representing the world—what we philosophers call intentional content--until, exasperated at the difficulties, some of the most influential among them announced that sensory experience just intrinsically represents the way things are. It’s what’s made these philosophical defenders of the manifest image’s personal level ((® Dan Dennett)—made them resist your user-illusion argument: Who’s the user? An illusion requires a conscious subject who’s wrong about something, who thinks wrongly that it’s not illusory. And the truth or falsity of that sentence the user entertains (“It’s an illusion”) is enough for them to treat subjective meaning (non-conceptual content) as undeniable—whence the attachment to the manifest image, in spite of 400 years of scientific unraveling.

You know the unraveling. Between them physics and physiology have unmasked much of the manifest image as illusory: color isn’t out there in the world, just photons changing the shape of rhodopsin molecules in our retinas, sending electrochemical signals to the visual cortex. That’s where the color is, if it’s anywhere. No solid, smooth, cool marble table tops either, just empty space in which subatomic particles move at half the speed of light, repelling the like charges in the molecules of our fingertips. The user illusions of color and solidity are parts of a quick and dirty solution put into place by the Darwinian process for all metazoans well before it foisted meaning and purpose on us, the last couple of quick and dirty solutions to the special problems we faced—particularly the organization of cooperation and collaboration. We know the saying, “Good enough for government work.”

I think you agree with much of that. You write,

“The key move is to recognize that we have underprivileged access to the source or cause of our convictions about what we experience… The idea that our being thus informed about these familiar things is caused by our being somehow directly or immediately acquainted with “inner” mental properties is just a bad theory that glosses over our ignorance of how this is done.”
It’s not even underprivileged access. We have no direct epistemic access to the source or cause of our convictions—the nonconscious cognitive processes in our brains that furnish us the qualia mistaken by the manifest image for accurate awareness of the world and mistaken for the material of representational thought. Science has no trouble identifying the causal role of the conscious play of felt experiences in the economy of human behavior. We’ll have to wait a bit longer for the full neuroscientific story of how the neural states constituting these experiences work their magic in detail. Meanwhile, we can already see plainly that the experiences don’t, can’t do the work of making meanings and purposes. They’re just more electrochemically constituted (hetero)phenomenology (® Dan Dennett).

You’re right to hold that saving what’s left of the manifest image of us, and of our lebens-world (phenomenology-talk for the manifest image), demands we naturalize reasons. The intuition-pump (® Dan Dennett) you use for that conclusion is hard to resist: “The eagle’s eye is a measurably better seeing thing for an eagle than a sheep’s eye or a fish’s eye would be, and there are reasons for that.” [Emphasis added] But “measurably better than” is not equivalent to what it is for reasons. And unless ‘reason’ is just a synonym for cause (which we both know it can’t be), you’re going to have to show how the process of blind variation and environmental filtration gets converted into a justification for the measurably betterness of the eagle’s eye. If you could do that you’d have constructed the manifest image’s ought (and its symbolic content) out of a natural historian’s is. You ask rhetorically, “There can be no serious skepticism—can there?—about what an eagle’s eyes are for.” I’m afraid there can be, or biology would be prepared to read functions off of nature with Bishop Paley’s abandon. I am reminded of J.B.S. Haldane’s (apocryphal?) answer to the question why are there 300,000 or was it 3 million species of beetle: “The creator’s inordinate fondness for them.”

Your peroration about Quine’s indeterminacy of radical translation makes manifest the practical, instrumental, nature of the notions foisted on us in the environment of evolutionary adaptation--“purpose” and “meaning,” and the other notions Homo sapiens constructed from them, named by words like “action” and “agency,” “desire” and “belief.” We both agree that Quine’s theoretical point is sound. There could in principle be two completely adequate incompatible translations from one language into another. But, as you also rightly note, “The odds against there being another good decryption or translation are astronomical….” Then you continue, “[W]e can rely on this practical impossibility and not try to cement it in place with some imagined internal fact that settles the matter of the right meaning once and for all.”

So, here’s the disenchanted or disabused naturalists’ last question (and their answer to it): Why is it we have to settle for practical impossibility and not theoretical impossibility when it comes to meanings? The question of why we have to settle for less than science demands and less than it gets everywhere else, is especially pressing when we have science’s explanation of why the impossibility of multiple translations is practical, an explanation of why meanings work for us, at least up to unique-for-all-practical-purposes decryptions. The explanation is the Darwinian one of the emergence of just good-enough survival tools, ones whose limits science reveals.

You conclude that thanks to scrutability of translation the language we share allows us to turn the representation of reasons into causes with great reliability. But the trick you need to perform is showing how mother nature by herself converts mere causes into the representation of reasons. That’s the hard challenge faced by anyone who wants to keep the manifest image afloat in the rising tide of science.

Alex Rosenberg

Alex Rosenberg

We do agree about most of these issues, and that leaves us in a vocal but growing minority among philosophers. I am happy to count you as an ally in the campaign to show philosophers how to distrust their mind-first perspective. After centuries of reflecting and “introspecting” and assuming (with Descartes and many others) that they have some kind of privileged access to their own thoughts and what their thoughts mean, these philosophers need a wake-up call. They need to undergo yet another “strange inversion of reasoning” (Dennett, 2009, 2013) if they are to abandon this fantasy and acknowledge that their own confidence in their knowledge of what they mean is based not on any “direct” acquaintance with meanings or purposes, but rather on a grand practicality: we almost always find that we can accomplish fruitful agreement on meanings by comparing notes, interpreting each other, and then interpreting ourselves with the help of those who engage with us. It’s a giant benign conspiracy of consilience, challenged, tweaked and improved, but lacking any “foundations” in bedrock facts of physical science.

As Quine so often said, paraphrasing Neurath, we are like sailors who must rebuild their ship while keeping it afloat. To my mind, the most persuasive and detailed positive account of this conception of meaning is Ruth Millikan’s, in Beyond Concepts: Unicepts, Language and Natural Information (2017), which handsomely completes her campaign—launched in 1984 in Language, Thought and other Biological Categories—against what she called meaning rationalism.

I see nothing in the vision I share with her that you should find objectionable, were it not for your unshakeable yearning for reduction to microphysics. You say “We’ll have to wait a bit longer for the full neuroscientific story of how the neural states constituting these experiences work their magic in detail.” You’ll have to wait forever, I think, if you tie both hands behind your back and demand that the full neuroscientific story eschews all use of intentional idioms. That is the message that Michael Levin and I (2020) have been urging on biologists in general, not just neuroscientists.

“Do not anthropomorphize genes, they do not like it.” (Yanai and Lercher, 2020)

This joke nicely needles the overreactions to excess anthropomorphism among biologists. As Richard Dawkins showed us almost fifty years ago, we can cash these terms out, and since we can, we may help ourselves to a little “deficit spending,” taking out intelligence loans that we can pay back later, once we get a better picture of the whole system (Dennett, 1971). Yanai and Lercher propose a distinction between day science (which shuns the intentional stance) and night science (which is when the new hypotheses get hatched for testing), but as long as we keep track of our obligations, we can avoid theoretical bankruptcy without making this drastic divide.

The economic metaphors here are particularly apt. Think of the equity in your house. Is it a real thing? What does it reduce to in terms of microphysics? It is real enough to be worth taking very seriously in its own terms. These terms include a crisp distinction between signal and noise, one of the most fruitful conceptual innovations of the last century, since it permits us to appreciate that patterns don’t have to be noise-free to be real (Dennett, 1991). Alex, you know much more economics than I do, and you surely recognize that economic realities are tricky but manageable. You can do real science in economics, but you can’t reduce it to microphysics—so what? When I see you worrying that biologists, if allowed the intentional stance, will “read functions off of nature with Bishop Paley’s abandon,” I wonder if you are still under the influence of Gould and Lewontin’s (1979) misbegotten attack on adaptationism.

You close with a challenge: “But the trick you need to perform is showing how mother nature by herself [my emphasis] converts mere causes into the representation of reasons.” Actually, that’s a very good summary of what I take myself to have done in From Bacteria to Bach and Back: The Evolution of Minds (2017). Here’s how I would put it in drastically foreshortened summary today: first mother nature (natural selection) converted causes to reasons (“how come?” to “what for?”) in the living world, uncovering millions of “free-floating rationales,” reasons that existed billions of years before there were reasoners, and then mother nature grew one branch (so far) on the tree of life that evolved language, and thus were reasoners evolved, by a coevolution of genes and memes that created people that would eventually become curious about all the how comes and what fors, and thus we now have the representation of reasons, one of the most powerfully fecund fruits of the tree of life, all accomplished without skyhooks.

Daniel Dennett

Daniel Dennett

Do you really want to admit that our confidence in meaning and purpose is based on “a grand practicality,” a “giant benign conspiracy,” lacking any bedrock “foundations” in physical science? All I’d like to add to that concession is the observation that science can explain why it’s at best a practicality and what it actually gets wrong about the nature of reality.

You’ve been rightly extolling the ‘teleosemantics’ Ruth Millikan has been working at for more than 30 years. Generous of you, as the idea was first expounded in your Content and Consciousness (1969). The core insight of teleosemantics is that neuroscience can give neural circuits something very close to meaning by showing how they are shaped by selection operating at the scales of evolution, development and behavior (Dennett’s Darwinian, Skinnerian, Popperian creatures from my first letter). If this approach, that Millikan has done so much to advance (along with the late Karen Neander), succeeds it will have successfully completed the program of expunging purpose from nature that Newton began. To preserve the manifest image as a description of reality, instead of a grand practicality it would have to do a little more. It would have to do the same job it does for neural circuity for the manifest image’s person level. That’s what most teleosemanticists reckon can’t be done.

You don’t think it needs to be done. As you put it in your last letter. “We may help ourselves to a little ‘deficit spending,’ taking out intelligence loans that we can pay back later, once we get a better picture of the whole system (Dennett, 1971).” Then you invoke a couple of nice metaphors: First, Yanai and Lercher’s (2020) distinction between day science (which shuns the intentional stance) and night science (which is when the new hypotheses get hatched for testing), but as long as we keep track of our obligations, we can avoid theoretical bankruptcy without making this drastic divide.”

You’re within your scientifically underwritten rights of course to take out the loan. The reward structure of science encourages any short cut to a new testable idea. There is no logic of discovery, anything goes in the process of what Einstein called “free creation.” We can help ourselves to as much of the “design stance” or the “intentional” one as we want. There are no more limits on how we enlarge scientific knowledge than there are on how we enlarge modern art. The difference with art emerges only once the act of creation is over and the work is subject to Baconian torture—experimental test.

The second metaphor, the economic one is, as you say, “particularly apt. Think of the equity in your house. Is it a real thing? What does it reduce to in terms of microphysics? It is real enough to be worth taking very seriously in its own terms.” The comparison of intentional psychology to economics is more than apt. Economics studies social constructions, institutions like exchange, the money that enables it to proceed, the interest rate that enables us to agree to exchange current consumption for greater later consumption, etc. There is no reality of these institutions independent of the manifest image. Seeing that is enough to forestall the temptation to reduce them to behavior or biology, still less to mind-independent physical reality. But just like economic institutions, the conceptual institutions that the intentional stance employs are social constructions, long ago made indispensable in the path dependent trajectory of human biological and cultural evolution. Does that accord them enough reality for you? I don’t think so.

The key word in our argument is ‘representation.’ I just don’t think you are taking that word seriously, as conveying the state of ‘presenting something else’—that’s what the ‘re’ does in representation. Reasons have to represent. That’s because they’re justifications that work only in arguments (sometimes one premise arguments). The reason why, the what for has content, a statement (one that we need a sentence to express), and that’s what it represents. Now I see how we represent in public language, turning inscriptions and noises into symbols. I don’t see how, prior to us and our language, mother nature (a.k.a Darwinian natural selection) did it. For that’s what she’d have to do in making causes into reasons, turning signs of what’s coming into symbols of what’s coming.

One more chance to show me where I’ve gone wrong?

Alex Rosenberg

Alex Rosenberg

Closer and closer. You write “There is no reality of these institutions independent of the manifest image.” I agree, of course. There is no reality to colors either, or aromas, independent of the manifest image. So what? You say “Does that accord them enough reality for you? I don’t think so.” But that is enough reality for me.

You must remember that the scientific image was nagged into existence by perplexities in the manifest image. How do magnets do that? How does a siphon or a gyroscope or an eclipse work? Our ancestors were not puzzled by many of the features of the manifest image which mother nature has fashioned for us. A prescientific farmer could give a pretty good explanation of why a ceramic mug shattered when dropped on a stone floor, how a pair of oars could propel a boat. But science soon uncovered puzzles and mysteries galore, and has gone on not only to solve them, but to use the acquired knowledge to enlarge and enhance the manifest image with boats that can sail upwind, telescopes, radios, computers, and all the high tech regularities our grandchildren take for granted. And in every case, we have acquired good hard empirical knowledge of reliable patterns in nature, and the proof that these patterns are “real enough” is that we can make big mistakes when we misuse them.

The manifest image is obdurate, not compliant with our wishes. Which brings us back to Dr. Samuel Johnson kicking the stone. Some philosophers have called this the argumentum ad lapulum, (the argument from kicking a stone), declaring it a logical fallacy. Well, maybe in some of its applications it is a fallacy, but it’s still a pretty good reality test, then and now.

One more thing. You say: “I don’t see how, prior to us and our language, mother nature (a.k.a Darwinian natural selection) did it [represented anything]. For that’s what she’d have to do in making causes into reasons, turning signs of what’s coming into symbols of what’s coming.

I don’t think you appreciate how radical my line on reasons is. Mother nature didn’t represent reasons at all; nothing did and nothing had to. We humans are the first reason representers on the planet and maybe in the universe. Free-floating rationales are not represented anywhere, not in the mind of God, or Mother Nature, and not in the organisms who benefit from all the arrangements of their bodies for which there are good reasons. Reasons don’t have to be representations; representations of reasons do.

The same is true of numbers of course. Long before there were people, let alone August Kekulé, benzene had a six-sided structure. Nobody appreciated that fact until very recently, but it has been a fact for billions of years. Bears have a good reason to hibernate, but they don’t represent that reason, any more than trees represent their reason for growing tall: to get more sunlight than the competition.

Alex, this has been a stimulating exchange. I understand your view better and you understand mine better, because, after all, they are largely the same view, differently emphasized.

All the best.

Daniel Dennett

Daniel Dennett