Sunday 8 July 2012

Gualtiero Piccinini: Is Consciousness a Spandrel?


      Abstract: Assigning a biological function to phenomenal consciousness appears to be needed to explain its evolutionary origin. For evolution by natural selection operates on organisms' traits based on the functions they fulfill. And yet identifying the function(s) of phenomenal consciousness has proven difficult. Some have proposed that the function of phenomenal consciousness is facilitating mental processes such as learning or reasoning. But mental processes such as learning and reasoning seem to be possible in the absence of phenomenal consciousness. It is difficult to pinpoint in what way phenomenal consciousness enhances such processes. In this paper, we explore a possibility that has been neglected to date. Perhaps phenomenal consciousness is a spandrel, that is, a byproduct of other traits that has no functions of its own. If so, then phenomenal consciousness has an evolutionary explanation even though it fulfills no biological function.

      S. J. Gould and R. C. Lewontin (1979), 'The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme,' Proc. R. Soc. Lond. B 205, 581-598.

Comments invited

18 comments:

  1. SPANDRELS, ACCIDENTS AND THE BIG BANG

    It's easy to find plausible "spandrels" or accidents, for structures and functions (doings). *Much* harder for feeling itself.

    If feeling is a spandrel (or accident), it's a **big** one -- maybe the biggest one since the Big Bang...

    ReplyDelete
    Replies
    1. I'm curious about why you think that feeling is a big deal (not that I disagree). It doesn't seem to make any difference in the world whether we feel or not. Maybe you have a bias to think that feeling is important because you are a "feeler", the only being to whom your feelings matter?

      Delete
    2. Feeling is interesting precisely because it doesn't make any difference in the world whether we feel or not, and yet we do. And so the question is: why?

      Delete
    3. Actually Stevan, I quite like the idea of consciousness as "a big [big big] accident". But there is an even bigger accident since the Big Bang... Isn't life the greatest accident on earth? why consciousness, why life? same questions! accidents happens and if we cannot find any adaptive function to consciousness, so why exclude the possibility of seeing consciousness as an accident?

      Delete
    4. I agree with you Pauline. So many random things put together lead to life. Maybe it's not a "mistake", but it was totally random! As to consciousness, Piccinini said that "all we can do with consciousness we can do without consciousness", he just said that without giving much examples. So many things wouldn't have been invented without consciousness, art, love, ect. Would altruism exists in species that we expect not to have consciousness? So I'm not convinced about the idea that our consciousness have no function therefore it's a spandrel, I think there is much more studies needing to be done about that.

      Delete
    5. "Isn't life the greatest accident on earth? why consciousness, why life? same questions!"

      I disagree. Many of the questions we are asking during this Institute are rooted in an implicitly accepted theory of evolution, whereby selection is guided by adaptive function. The question of "why life?" pertains to an altogether other domain of inquiry.

      Delete
  2. Stevan, I agree that if feeling is a spandrel or accident, it's a big one, in the sense that feeling is a big deal to us. it's the most precious part of us; without it, in an important sense, there wouldn't really be an "us"!

    But then again, as Martha points out, it's not clear that feeling makes any difference. That's where I'm coming from. Personally I think and hope feelings makes a difference, but I find this an extremely difficult thing to establish.

    ReplyDelete
  3. Argument from epiphenomenalism: GUALTIERO PICCININI
    1)epiphenomenalism says that consciousness is causally inert.
    2)Causally inert properties cannot be selected for (for lack of effects on which natural selection can operate)
    3)Therefore, consciousness is either a spandrel or an evolutionary accident.

    But, if you adopt supervenience:
    1) the physical properties, that are the supervenience base of epiphenomenal properties, ARE causal (they cause other physical events)
    2) the physical properties thus DO have effects on which natural selection can operate
    3) the physical properties (the supervenience base) can be selected for
    4) the epiphenomenal properties that superviene on them, come with them (selected indirectly), despite not causing anything themselves

    ReplyDelete
    Replies
    1. I should add that that this argument I just gave is intended to show that consciousness (if epiphenomenal) could indeed still be a spandrel, but could not be considered an evolutionary accident.

      Delete
    2. Pietro, thanks for your comment. The problem with your argument is that not all physical properties of an organism are selected for. Some of them may be acquired accidentally, e.g. by genetic drift. If phenomenal consciousness supervenes on those, then it's an accident.

      Delete
  4. I agree with Dr. Piccinini that there are a lot of theories but little (if any) evidence to really show that p-consciousness serves any function. I concur and believe (maybe because I am a neuroscientist) that it is a neural event, a physical thing. It does not seem readily apparent to me how one can show that, but I guess one will never find out unless one tries!?

    ReplyDelete
  5. The argument that all the living organisms can do consciously (and especially what humans can do consciously) could also be done without consciousness is probably a mistake that might be the result of a misunderstanding of the principles underlying evolution, especially if we see consciousness as a spandrel. In fact, one of the important principle in evolution is that evolution works with what already exists. So if consciousness appeared by mere chance at some point in evolution but was completely neutral with no reason to be "selected against" or eliminated by natural selection and that for some other reasons (such as the fact that it appeared to be a characteristic found in individuals that developed another adaptive trait or by mere genetic drift for instance) it remains through the evolutionary time, evolution had to cope with this "parasitic" feature that could eventually be used to increase the adaptive properties of other adaptive functions that do not actually necessarily need consciousness to be a minimum adaptive.

    ReplyDelete
  6. Watch out Dr. Piccinini, there are retinal neurobiologists in the audience. Your analogy of retinal architecture as an evolutionary spandrel has a serious oversight...
    It is now well understood that the mammalian retina is arranged 'inside-out' due to the presence of both the retinal pigmented epithelium (RPE) and intrinsically photo-sensitive retinal ganglion cells (ipRGCs). ipRGCs (a special subclass of retinal neuron) sit closer to the light source than traditional rods and cones, in order to absorb ambient light in the red spectrum. These cells drive both circadian rhythms and the pupillary light reflex. If RGCs were to sit at the back of the retina, pointing their axons away and absolving the need for a blindspot, our rods and cones (and the RPE that comes along with it) would absorb the portion of ambient light required by ipRGCs to drive the aforementioned low level visual responses.
    The RPE, located at the back of the eye, is absolutely essential for human colour vision in it's ability to exquisitely regulate the behaviour and survival of rods and cones. We could't have the RPE anywhere else on the eye, because, well, it would block all light - then we really wouldn't be seeing very well at all...
    We compensated for this seemingly 'inverted' arrangement by evolving binocular vision. Yeah, if we loose an eye we might have some problems in the visual field. Coincidentally, if we loose a lung, we might have some problems breathing. In the words of WakeupSheeple1111, 'Duh.'.
    Do you have any other salient examples of evolutionary spandrels?

    ReplyDelete
  7. ATufford, thanks so much for your helpful comment. Based on what you say, the hole in the vertebrate remains a spandrel, i.e. a byproduct of the arrangement you describe. But the arrangement you describe, it seems, is not an accident, contrary to what I suggested in my talk (although it's still not clear to me why the axons have to point towards the eye; I'm just going to assume that there is a reason for that). So what I need is a different example of evolutionary accident. Anything generated through genetic drift will do.

    ReplyDelete
  8. Even if we are satisfied that feelings just happened as a big, useless accident (I'm not), it doesn't explain how doings become feelings. There is an explanatory problem there whether or not we believe feelings confer an adaptive advantage.

    ReplyDelete
  9. During the question period, Diego Mendoza raised the following objection: all cases of evolutionary by-products are simple and since phenomenal consciousness is complex, it would be extremely unlikely that it is a spandrel. I think there was really something to his argument even though using the word 'complex' might have weakened his point a bit. (Dr. Piccinini simply denied that p-consciousness was complex.) Let me try another formulation: there are two aspects of p-consciousness that seem to suggest that it is something highly organized that could neither be a by-product or an accident: (1) Our p-conscious perceptual states are well-integrated across sensory modalities: if we see a pig in front of us, we hear it grunting, we smell it and we touch it, we perceive these features as belonging to the same entity, not as five different, unrelated features of the world. (2) Our p-conscious mental states often carry information that is needed to plan our actions over longer periods of time (which is a process that forces us to integrate very different kinds of data -- e.g. our knowledge of weather conditions, of other people’s intentions, of the danger associated with certain decisions). On the other hand, they usually carry little information regarding the internal states of our organs -- except when one of our organs is severely damaged and we feel pain. For instance, our p-conscious perceptual states carry information about the objects lying on your path, the people that surrounds you and the color of the sky, but not about the approximate number of white cells you have in your body or whether certain neurotransmitters have been emitted in your brain. The content of our p-conscious mental states are thus often used to make incredibly complex decisions about our future that requires us to weigh an indefinite number of factors against one another. Yet, when it comes to serious issues about our bodies (which require our brain to weigh a specific set of pre-determined factors against one another), the state of our internal organs is not part of the content of our p-conscious mental states.

    ReplyDelete
  10. I found Dr. Piccinini's talk really interesting but I can't help but have a difficulty with the concept of spandrels... I find that saying that something is a spandrel, an accident or a byproduct of something comes up only when we don't find any compelling story for a trait or feature. An absence of evidence is not an evidence of absence and I find that such evolutionary explanations fall easily in this trap,

    ReplyDelete
    Replies
    1. There is another trap: coming up with a seemingly convincing or "plausible" story for a trait that turns out to be a "just-so" story.

      Take, for example, jealousy. David Buss comes up with a plausible story about how jealousy is a natural and evolved trait. I'm overly simplifying his argument here, but basically the idea is that because of sexual competition and because males "invest" in a mate, they are perfectly rational when they're being jealous. This of course can be read in a number of ways, one of which is that it legitimizes jealousy-based behaviors, "because it,s just how we're wired". But really, this whole argument is based upon assumptions about the environment of evolutionary adaptedness. And yes, we need to make those assumptions at some point, but they need to be backed up by empirical evidence, lest we find ourselves trying to decide which story is the best based on their apparent plausibility. The point here is that trying to find a plausible story for any and all traits humans exhibit can lead to methodologically unsound proposals.

      Another point to consider is that saying a given trait is a spandrel does not amount to saying it doesn't have a function. There are numerous different selection processes at work in evolution, one of those being exaptation, by which a trait is co-opted to fulfill another function than the one it was selected for OR by which a trait with no prior function acquires one.

      So, saying consciousness is a spandrel does not mean it has no function.Furthermore, assuming it is a spandrel BECAUSE so far we haven't found a plausible adaptive function for it will enable us to avoid unsound or uninteresting evolutionary explanations. evolutionary explanations are very powerful, but they do come with a number of methodological problems.

      Delete