“Purpose” in science and morality

IMG_0018A New video in the ongoing Kaufman-Pigliucci series is out, this one on the question of whether teleology, the idea that things have a “purpose” in the strong Aristotelian sense of the word, still makes sense in light of modern science and philosophy.

We begin our discussion by examining various meanings of “purpose” in science and in morality, and then by exploring Aristotle’s take on the subject. I argue that “what is it for?”, i.e., looking for functions, makes perfect sense in evolutionary biology, but not in other sciences, such as chemistry or geology. That’s because of the special role of natural selection in evolution. Accordingly, we explore the relationship between form and function and how the two reciprocally shape each other in living organisms.

We then move to ethics, exploring the idea of moral laws. From there, we discuss the different paths to human flourishing and how they relate to the concept of meaning and purpose. Finally, I explain once again what Sam Harris gets wrong about the relationship between science and ethics. But you can skip that bit if you are (understandably) tired of that particular dead horse… (or you can read my original critique here).

Here is the full video:

156 thoughts on ““Purpose” in science and morality

  1. Daniel Kaufman

    Massimo wrote:

    I think the difference is that the workings of natural selection are directional, cumulative, and tend to maximize particular quantities (survival and reproduction).

    = = =

    Yes, I can see this. One more thing. Are directionality, cumulativeness, and being a particular maximized quality in any way properly seen as intrinsic or are they ultimately a function of human points of view and frames of reference?

    Liked by 1 person

  2. Bunsen Burner

    Massimo:

    ‘I’m suspicious of anything that is cast in formal terms of necessary and sufficient….’

    Hence gray areas such as a virus or the progenote. Of course some people have different ideas as to where even the gray areas are, for example Biospheres or computer life. It seems to me that it has more to do with the utility of the explanation to us as human beings than anything else. Not that I think there is anything wrong with that.

    Liked by 1 person

  3. Massimo Post author

    Dan,

    I would say they are intrinsic properties, independent of a human frame of reference. Biological evolution, obviously, has been going on, pretty much in the same way as it proceeds today, for billions of years before we came on the scene, and while the collection and interpretation of the pertinent data is, of course, a human endeavor, subject to human bias, it is no more so than the criteria by which astronomers collect data on stars and galaxies.

    Like

  4. Coel

    Hi Richard,

    … emotivists believe that the utterance of a moral statement is JUST an expression of emotion, and not an assertion of a supposed fact. But I say that people who utter moral statements in earnest (e.g. “Abortion is morally wrong!”) are very often (but perhaps not always) asserting what they take to be a fact.

    Agreed. And thus (perhaps like you?) I subscribe to both emotivism and error theory. Most people are intuitively moral realists, and hence human language is usually moral realist. Thus when (most) people make moral assertions they think that they are asserting moral facts. But they are wrong. What they are actually doing is reporting their own feelings on the matter.

    Like

  5. Coel

    Dan,

    The idea, then, seems to be that natural selection is a kind of system, while, as Massimo explained, the the geology of the planet is not. […] Why do these facts make natural selection a “system” any more than the entirely accidental fact that certain combinations of natural elements like water and soil and rock cause certain things to happen and others not to happen?

    I’d suggest that the difference is that natural selection is a directional process, one that can be thought of as finding a (local) maximum of a “fitness function”. It is that directionality that then makes it appropriate to adopt talk of “purpose” in biological systems.

    In contrast, geological cycles and weather cycles can be regarded as “systems”, but they do not have anything corresponding to the directional optimisation of a fitness function.

    Liked by 1 person

  6. Daniel Kaufman

    Massimo: This is what I don’t quite understand. How can something like directionality be intrinsic? Certainly evaluative directionality — i.e. ‘better’ and ‘worse’ and all the variations upon — are relative to human points of view and frames of reference, but even purely descriptive directionality would seem to be as well — i.e. ‘upwards’. ‘downwards’, ‘forward’, ‘backward’, etc.

    ‘Cumulative’ seems straightforwardly analogous.

    As for maximizing certain properties, isn’t the decision of which properties, the maximization of which to focus on, is a matter of human decision? I thought this was part of the point of Fodor/Piatelli’s book (and I’m only pointing to this one, lone feature, not by any means intending to open discussion on the book more generally.)

    Liked by 2 people

  7. Philosopher Eric

    Massimo,
    It seems to me that ordinary definitions for the term “purpose”, require a subject of reference in all cases, such as my own entertainment. Saying that evolution has an intristice purpose, unlike anything else, is thus problematic in a definitional regard. Consider the possibility that you have instilled evolution with the purpose that you now perceive it to have. I think you’ll find it difficult to sensibly maintain an inherent purpose — that is unless your talents trump our lack thereof. In my view attempting such an argument would be far more problematic than, for example, arguing a good case for moral realism.

    Like

  8. Massimo Post author

    Coel,

    When I say that not taking care of the future of the planet in order to cater to special interests is wrong I am not expressing (only) an emotional state. I am making a conditional prescriptive statement: IF we give a shit about our future and the future of our kids THEN we ought to do certain things. There is no error here, nor is this just a matter of subjective emotional response.

    Regarding selection, no, the directionality isn’t sufficient. Plenty of other natural phenomena are directional, e.g., erosion. It’s the existence of a fitness function that is the key difference.

    Dan,

    Fodor and Piattelli really got it wrong in that book, see my review, for instance: http://tinyurl.com/ybg5vdkv

    The fitness function of a given population of organisms does not depend on a human frame of reference, only on the specific environment(s) inhabited by that population, together with the phenotypic and genetic characteristics of the population in question.

    Eric,

    I did not say that evolution has an intrinsic purpose, only that it generated organisms (us) capable of acting purposefully. There is a huge difference.

    Like

  9. synred

    Occam’s razor is not a magic wand by which to wish away all philosophical problems, and you should stop using it that way.

    Occam’s razor is just a ‘rule of thumb’. Sometimes, quite often, simple is not sufficient. Esp. in biology.

    Like

  10. Massimo Post author

    Dan,

    No, I don’t. Directionality is very common in natural processes. For instance, the average entropy of the universe keeps increasing. Nothing to do with a human framework. Same for cumulativeness: rivers cumulatively add sediment to their bottom because of erosion. Again, no need of a special human framework.

    Liked by 1 person

  11. Coel

    Hi Massimo,

    I am not expressing (only) an emotional state. I am making a conditional prescriptive statement: IF we give a shit about our future and the future of our kids THEN we ought to do certain things.

    Yes agreed. But such instrumental-ought statements are, in themselves, purely descriptive and morally neutral. They are the things that science can do. To get a moral prescription one needs to add in a human value or desire.

    Liked by 1 person

  12. Massimo Post author

    Coel,

    Conditional imperatives are not prescriptive, nor are they morally neutral. And no, science (by itself) can’t do them. But I will terminate this discussion here, out of fairness to DM, following Dan’s reminder that this isn’t really what we are talking about.

    Like

  13. Philosopher Eric

    Eric,
    I did not say that evolution has an intrinsic purpose, only that it generated organisms (us) capable of acting purposefully. There is a huge difference

    Well that’s good news Massimo! What would you say separates organisms like us that can act purposefully, and organisms that can’t?

    Like

  14. synred

    DM: Why don’t you write up your “idea” for refuting moral realism via Occam’s Razor and send it to a peer reviewed ethics journal?

    I find a difficult to see how Occam can defeat any idea. Sometimes the ‘simple’ explanation is not correct. Simple is a good place to start, but that’s all. It doesn’t prove anything; it’s not evidence.

    Like

  15. synred

    like you focus on the subset of mathematical objects that are coherent

    I don’t see that coherence is completely neccessary. I’ve written lots of incoherent programs that run fine for awhile. Until they crash or spit out the wrong answer.

    Like

  16. Coel

    Dan,

    Certainly evaluative directionality — i.e. ‘better’ and ‘worse’ and all the variations upon — are relative to human points of view and frames of reference, but even purely descriptive directionality would seem to be as well — i.e. ‘upwards’. ‘downwards’, ‘forward’, ‘backward’, etc.

    The issue of how many descendants one has in succeeding generations is a matter of objective fact, and not dependent on human view points.

    As for maximizing certain properties, isn’t the decision of which properties, the maximization of which to focus on, is a matter of human decision?

    In artificial breeding, yes. The dog breeder or pigeon fancier can choose which properties to breed for. But “natural” selection — the fact that things that are better at leaving more descendants will leave more descendants — occurs automatically and necessarily. That is indeed why Darwin called it “natural” selection (after spending the opening chapter of OofS talking about artificial selection).

    Liked by 2 people

  17. synred

    It seems to me evolution may only be progressive in retrospect (e.g., see Gould). If teleology is a result of progressive evolution the argument may be somewhat circular.

    Liked by 2 people

  18. synred

    Fitness functions typically have multiple local maxima. Which you end up on has some element of randomness, hence you can’t ‘rewind the tape’ and expect to end up in the same place. And any way fitness function only gets you teleonomy.

    Even thermodynamics shows some teleology. A closed system “wants” to come to thermal equilibrium, though all that’s really happening is statistics of large numbers mean the most common configurations dominate — by a lot — no teleology needed.

    Liked by 1 person

  19. Thomas Jones

    I listened to the broadcast and was most interested in trying to get a better handle as a layperson on teleonomy and teleology and why this might be important to me. I.e., I don’t envision having to make such distinctions in most everyday discussions unless confronted by an ID argument, or–more distressing to me–discussing purpose and cause with a gun rights activist who reverts to the NRA’s slogan, “Guns don’t kill people, people kill people,” both of which I try to avoid in the interest of preserving some sense of equanimity.

    The current Wiki article on teleonomy ends with this statement:

    “Teleonomy is closely related to concepts of emergence, complexity theory,[13] and self-organizing systems.[14] It has extended beneath biology to be applied in the context of chemistry.[15][16] Some philosophers of biology resist the term and still employ “teleology” when analyzing biological function[17] and the language used to describe it,[18] while others endorse it.[19]”

    Allen MacNeill on his blog discusses Ernest Mayr’s distinction, but states parenthetically:

    “I would state this slightly differently from Mayr: that there is no observable evidence that the evolutionary processes by which such programs come into being are goal-directed (i.e. “designed” or “purposeful”). Therefore, although such purposes may exist, they are invisible to us on principle and therefore irrelevent [sic] to scientific explanations of natural phenomena.”

    If the distinction is irrelevant to scientific explanations of natural phenomena, why the interest in it? The use of either term without a predetermined context seems indicative of a lack of clarity or laziness in exposition more than anything else. Why not avoid the usage of either in such cases?

    Liked by 1 person

  20. couvent2104

    I’d suggest that the difference is that natural selection is a directional process, one that can be thought of as finding a (local) maximum of a “fitness function”.

    A minor point, but quite a few physical laws can be formulated as consequences of variational principles (principle of least action, the path of light through different media etc.) which means that the dynamics of these systems are in certain precise way an extremum (minimum or maximum) of a function etc. One can derive classical mechanics from such a variational principle. Yet I’ve never heard somebody call classical mechanics “directional” (it’s invariant under time reversal).

    The only scientific way to call a process directional is to give a precise definition of a property X, give a precise definition of an “order” for X and prove that the evolution of X always has the same order.

    To give an example: if X is a property of a system that can expressed as a number; if X1, X2, … are subsequent values of X while the system evolves; and if you always have X1 < X2 < X3 < X4 < … and so on, then it’s reasonable to say that X has a direction. Not everything can be expressed as a number, of course, but there are other types of order possible.

    But even then the work isn’t done. It’s extremely easy to define properties of specific systems in classical mechanics that are ordered in this sense. Just take a ball that rolls away and define X as the distance between you and the ball. But we still don’t call classical mechanics directional. For that, we should be able to define an X on the general level of the laws of classical mechanics.

    In the same way (and in my personal opinion) the fact that some processes in ET are directional in some circumstances, is not enough to call ET directional. To arrive at that conclusion, there should be a directional property X on the level of the “laws” of ET. This X should be a uniquely and precisely defined directional property of all systems in all circumstances studied by ET.

    Can this be done? I’m curious.

    Like

  21. Coel

    Hi Richard,

    You cannot sensibly do that, because they [emotivism and error theory] are mutually inconsistent views.

    Why so? In making moral pronouncements, people think that they are referring to objective moral facts, but are actually reporting their subjective emotional values. Why is that inconsistent? [If the issue is one of labeling, how should that stance be labelled?]

    Like

  22. Massimo Post author

    Eric,

    “What would you say separates organisms like us that can act purposefully, and organisms that can’t?”

    Consciousness. Imagine, some misguided people even dismiss it as an illusion…

    Synred,

    “It seems to me evolution may only be progressive in retrospect (e.g., see Gould). If teleology is a result of progressive evolution”

    I think I said directionality, not progress. There is no meaningful concept of overall progress in evolution, only local one, as in individual populations become more and more adapated to their environment. But then the environment changes…

    “And any way fitness function only gets you teleonomy.
    Even thermodynamics shows some teleology”

    Yes, no. Right, fitness function gets you teleonomy, not teleology. But thermodynamics gets you neither. It is simply an example among many of natural processes that have a direction, but it doesn’t produce function (teleonomy) and certainly doesn’t yield purpose (teleology).

    Thomas,

    I actually don’t think one needs emergence to get teleonomy. As I said, it’s the result of natural selection, which is not considered an “emergent” process, except in the most trivial of senses. And not all self-organizing systems display teleonomy (hurricanes). And the entry is entirely wrong about philosophers of biology “resisting” the term. Mayr never talk about design or purpose, only of function.

    The distinction is extremely relevant to scientific explanations. Without it, one couldn’t make sense of the distinction between a hurricane and a vertebrate eye.

    Couvent,

    I don’t think I said classical mechanics is directional. But average entropy increase certainly is.

    Coel,

    As I said, I’d rather not continue the discussion on metaethics in this thread. But I must say that your comment about emotivism and error theory being compatible shows that you have not read the pertinent philosophical literature… From the entry of a pertinent SEP entry (http://tinyurl.com/ya95qx85):

    “Non-cognitivism is a variety of irrealism about ethics with a number of influential variants. Non-cognitivists agree with error theorists that there are no moral properties or moral facts. But rather than thinking that this makes moral statements false, non-cognitivists claim that moral statements are not in the business of predicating properties or making statements which could be true or false in any substantial sense. Roughly put, non-cognitivists think that moral statements have no substantial truth conditions. Furthermore, according to non-cognitivists, when people utter moral sentences they are not typically expressing states of mind which are beliefs or which are cognitive in the way that beliefs are. Rather they are expressing non-cognitive attitudes more similar to desires, approval or disapproval.

    Cognitivism is the denial of non-cognitivism. Thus it holds that moral statements do express beliefs and that they are apt for truth and falsity. But cognitivism need not be a species of realism since a cognitivist can be an error theorist and think all moral statements false. Still, moral realists are cognitivists insofar as they think moral statements are apt for robust truth and falsity and that many of them are in fact true.”

    As you can see, emotivism is a type of non-cognitivism, while error theory falls under cognitivism. And the two are, as the SEP says, the denial of each other.

    Liked by 2 people

  23. richardwein

    Hi Coel,

    The issue of how many descendants one has in succeeding generations is a matter of objective fact, and not dependent on human view points.

    Yes. I think that’s at the root of why it’s useful to talk of function or purpose in the context of natural selection. Reproductive success is not just a property that an observer might happen to be interested in. It’s central to the process of natural selection. Given that objective criterion for “success” and “better”, the rest of the language of functionality follows.

    Eyes are in the first place for producing offspring, and their function of seeing derives from that.

    Like

  24. wtc48

    Synred: “It seems to me evolution may only be progressive in retrospect (e.g., see Gould). If teleology is a result of progressive evolution the argument may be somewhat circular.”

    Human analysis can detect the operation of feedback factors that would improve the performance of specific traits (e.g. eyesight in falcons), or conversely, retard the same traits under different conditions (e.g., also eyesight in cave-dwelling or bottom-dwelling species). But this is retrospective, as we are the only historians. Pattern-seeking is a mixed blessing for historians.

    Liked by 1 person

  25. synred

    A minor point, but quite a few physical laws can be formulated as consequences of variational principles (principle of least action, the path of light through different media etc.) which means that the dynamics of these systems are in certain precise way an extremum (minimum or maximum) of a function etc. One can derive classical mechanics from such a variational principle. Yet I’ve never heard somebody call classical mechanics “directional” (it’s invariant under time reversal).

    While the equations of classical mechanics are reversible, the solutions need not be. If I put a bunch of particles in the corner of a box and run time ‘forward’ they will spread out and fill the box. There is a direction.

    If I run time ‘backward’ from the specified condition the particles will also spread out to fill the box. There still a direction.

    The direction of time does not depend on the equations being time-asymmetric, but on the initial state being low entropy.

    QM is more complicated. In standard interpretation ‘measurement’ is intrinsically asymmetric though the equations are still time symmetric (at least if you flip charge and parity too). In many worlds you can run time ‘backwards’ only if you flip all the worlds. However, if you examined the memories of the creatures living in such a world they would still only remember the ‘past’ even if you were running them ‘backwards.’

    See ‘Causality’ here https://goo.gl/pLClDR

    Like

  26. richardwein

    P.S. I wrote: “Eyes are in the first place for producing offspring…” Perhaps It would have been better to say that eyes are in the first place for reproducing themselves. It’s the success of traits at reproducing themselves which is at the core of natural selection.

    Like

  27. Thomas Jones

    Massimo, “And not all self-organizing systems display teleonomy (hurricanes).” Exactly what went through my mind when I tried to apply the terms as a layperson but that I don’t think it accurate to say that hurricanes are self-organizing. Perhaps, that’s where my confusion comes in. Moreover, except for what I consider crackpot evangelicals who imputed teleological purpose to Katrina–presumably to wipe out the LGBT community in New Orleans–most people I talk to innately understand that the purpose in forming a Rotary Club is unlike a purpose ascribed to a hurricane.

    I will concede that you know more about Mayr than I; still, he’s quoted as redefining telemony, “It would seem useful to rigidly restrict the term teleonomic to systems operating on the basis of a program of coded information. (p.42)” But this alone doesn’t help me. What this seems to entail, after eliminating natural emergence and/or supernatural agency, is the ascription of teleology to self-directed purpose or goal design in some biological species, but not all.

    I don’t share your distinction when you state: “Without it, one couldn’t make sense of the distinction between a hurricane and a vertebrate eye.” Not really. I think most get this distinction, though it’s true that it’s “extremely relevant to scientific explanations,” otherwise, I wouldn’t rely on my contact lenses to correct my vision.

    Going back to Mayr, if we are to rigidly restrict the usage of such terms as telenomic to certain systems, we can be expected to know when to do otherwise. That is still unclear to me as is suggested by your reference to thermodynamics.

    Like

Comments are closed.