Randomness, the brain, free will, science, "pseudo-science," justice, and demarcation: a conversation
The following is excerpted, with permission, from email-exchanges associated with an Evolving Systems Project discussion of chance. Continued public conversation, to which others are invited to contribute, can be found in the on-line forum area below. Comments can be individually linked to using the numbers in parentheses (eg http://serendip.brynmawr.edu/exchange/evolsys/chance/cashmore#1).
Mike Sears, 17 June, 2010 (1)
I just thought that I'd pass along an article that, given our morning conversations, you might find interesting.
"It is widely believed, at least in scientific circles, that living systems,including mankind, obey the natural physical laws. However, it is also commonly accepted that man has the capacity to make “free” conscious decisions that do not simply reflect the chemical makeup of the individual at the time of decision—this chemical makeup reflecting both the genetic and environmental history and a degree of stochasticism. Whereas philosophers have discussed for centuries the apparent lack of a causal component for free will, many biologists still seem to be remarkably at ease with this notion of free will; and furthermore, our judicial system is based on such a belief. It is the author’s contention that a belief in free will is nothing other than a continuing belief in vitalism—something biologists proudly believe they discarded well over 100 years ago."
Abstract from "The Lucretian swerve: The biological basis of human behavior and the criminal justice system" by Anthony Cashmore, Proceedings of the National Academy, USA 107: 4499-4504 (2010)
Wil Franklin, 18 June, 2010 (2)
This article was well, very generative for me. Particularly the juxtaposition of the whistle and train metaphor with the observations of individuals with lesions in the brain that lead to "blind-sight" or conscious blindness with the unconscious ability to use visual information. Let me first explore the equivalences between the two and then end with a question regarding the apparent mis-representation of the metaphor.
Starting with the metaphor:
Let's assume the whistle of a train is "like" consciousness in that it is a signifier. Let us also assume the steam building in the engine is the driving force of the trains motion, similar to neuronal activity in the unconscious brain is a driving force of behavior (as per Figure 1, Cashmore). Then the whistle sounds when steam passes through it, corresponding to enough steam pressure in the engine to initiate movement. Remove the whistle (like the lesion in the brain of some individuals) and you remove the sound (consciousness) but not the underlying steam and concomitant action.
Cute, yes. But missing something that I think is significant.
If all brain function (conscious and unconscious) is simply different patterns of neuronal activity and unconscious patterns of neuronal activity can affect behavior, then who is to say that a pattern of neuronal activity that happens to be conscious cannot affect behavior. I'm not sure I am saying anything about free will, but I am wondering whether or not consciousness is only a signifier? It seems to me the brain is wired such that there is no such thing as a "whistle". The whistle metaphor breaks down because the brain is interconnected with multiple, multi-directional modules affecting one another. The whistle is a module with only a unidirectional connection.
Maybe a connection to Free Will:
Now let me assert that the conscious module of the brain can generate "counter-factuals" (can form neuronal activity patterns that correspond to states other than are currently formed in the unconscious modules of the brain). This ability to generate a conscious conterfactual may only be the ability to bring into consciousness a state that is in the unconscious... in which case all this collapses into just a whistle and engine problem (Figure 1C.) But if the conscious module of the brain can (in reacting to current unconscious states) form a state that is not currently present in the unconscious, then it seems to me a conscious idea/thought (read pattern of neuronal activity) might affect action, not simply seem to affect it in a post hoc fashion.
Or is that exactly what I just explained? My counter-factuals are undermining my assertion before I can even finish. Any help?
I don't think that there is anything that I disagree with in there. However, I think it would be useful if he brought in more recent ideas from philosophy of mind on "Extended Minds":
The notion described in Cashmore's Figure 1c is still simplistic and is the reason people are left continuing to look further for "free will". The brain is not just a conversation between conscious and the unconscious, but a rich interaction between the conscious, unconscious, and the world---including senses, and conversations with other brains. Capturing that with the E in GES (genes, environment, and stochasticism) doesn't do it justice. If I were to draw that figure, E would be a node at least as
large as consciousness and unconsciousness, and have double-headed arrows between both.
Anne Dalke, June 19, 2010 4)
I'd be interested in your reading of the article, Mike. I don't buy it. One of us suggested years ago that the intrinsic variability of our nervous system, plus the capacity to withhold an output until that variability yields something that satisfies us, is an operative definition (and actualization) of free will (for details, see
That works for me: clearly, we're not fully under our own control, but combining the intrinsic variability of the unconscious with the monitoring-and-editing function of consciousness can lead to some pretty satisfying outcomes. Then if you add in (per Doug--> Andy Clark & David Chalmers') notion of extended mind (located outside our individual "biological skin bag," in computers and notepads and especially in unpredictable, uncontrollable-but-monitorable interactions with other people), well then--
Alice Lesnick, June 20, 2010 (5)
My response to Cashmore relates to our discussion a couple of sessions back about if and how language/word choice matters.
But before I share them, I want to appreciate the different comments you all have made, and I suppose raise a meta-question that arises from their richness for me. I am interested in what Cashmore's account of us suggests for the education system, indeed, for the meaning of learning at all. I want to ask him: When I learn from, and enjoy, different people's responses to your paper, situated as they are in different people's experiences, am I acting as a mechanical force of nature, conscious automaton, or bowl of sugar?
Connected to this is my wish to know more about Cashmore's senses of the terms he uses. I don't think such important ideas can be discussed well unless our language for them is subtle and nuanced, and honest as to its reach and limits, and, preferably, to our own experiences and investments. 'Environmental history," for example, is such a broad term -- how does it map to diverse people's experiences and representations of that history? Can that history be "reflected" clearly and once and for all?
When Cashmore discounts "writings on free will" because they "lack molecular details," I want to ask why everything that is written has to reference the same givens, the same founding ideas/knowledge base/field. Why is such reductiveness cast as virtue? Why not consider as well the possibility that it could be limiting?
Paul Grobstein, June 20, 2010 (6)
Among the things that struck me about the article was the inclusion, albeit somewhat grudgingly, of "stochasticism" as among the "three forces that govern behavior", the other two being genes and environment. Yes, there continues to be an uncertain border between "deterministic but ill-mannered" (http://serendip.brynmawr.edu/bb/EncyHumBehav.html) and probabilistic in the sense of non-deterministic, and this needs further exploration. At the same time, it seems to me noteworthy that more biologists are acknowledging "a third thing" (http://serendip.brynmawr.edu/gen_beh/Dreams.html) significant in thinking about behavior, and biological phenomena generally. Cashmore's catalogue of some of the places where biologists find it important to acknowledge stochasticity is a significant contribution to overturning a popular sense that biologists attribute everything to a combination of genes and environment. And to our discussions of how to think productively about "fortune - the aleatory, chance ... tychism." That which is not fully understood should not be dismissed on that ground, and in this case "stochasticism" is being increasingly recognized as something needing to be acknowledged and brought within the sphere of both legitimate inquiry and legitimate explanatory frameworks.
Along similar lines, I was struck by the article as further evidence that a similar process is occurring with regard to "consciousnness." Forty years ago, consciousness was dismissed by most biologists as, at best, irrelevant to productive biological inquiry and, at worst, as a meaningless distraction from it. That Cashmore finds it obligatory to consider consciousness in relation to the problem of "free will," and to consider possible causal roles of consciousness in behavior (a fourth "force"?) will, I think be useful as we continue our discussions of the relation between brains and Turing machines. I will argue that Cashmore has adopted much too narrow a conception of the function of consciousness (and that in turn contributes to difficulties in making sense of free will in neural terms) Cashmore's paper also provides another good example of the importance of biologists (and other inquirers) not dismissing phenomena on the grounds that they aren't yet understandable within particular research frameworks ("formal systems"?) but instead holding open judgments about their significance in the long run.
Very glad to have Cashmore (and you) put "free will" on the table. My own intuition, as a biologist, is that explorations of "free will" will follow the same future trajectory as those of stochasticity and consciousness. I'm not impressed by Cashmore's documentation of a past association between "free will" and dualism/vitalism. The same historical association existed for both stochasticity and consciousness, and, for that matter, for life itself. Biologists didn't stop exploring life because of the historical association, and we now understand life better, if differently, than we did before, in a way that doesn't depend on either dualism or vitalism. I'm also not impressed by Cashmore's extensive quotations from, among others, distinguished biologists (more on this and the previous point below). That something is not currently understandable by distinguished people representing a particular research tradition is not evidence that it doesn't exist nor that it won't become understandable at some future time as that research tradition itself evolves.
Along these lines, it seems to me that Cashmore is curiously neglecting the history of the research tradition of biology. Yes, living systems contain "chemicals" that obey " the laws of chemistry and physics" (insofar as we currently understand these). But to assert that "as living systems, we are nothing more than a bag of chemicals" is to grossly misrepresent more than a century of biological research. "A bag of chemicals" identical to those of any living organism does not display any of the properties of a living organism. The latter depend instead on particular organized interactions of those chemicals and, in turn, particular organized interactions of those into larger organized assemblies, of those assemblies into still larger organized assemblies, and so forth at multiple levels of scale. The biological research tradition has been not one of denying the existence of things that can't be made sense of in terms of organized matter but rather one of progressively recognizing the ever growing array of sophisticated and often unexpected properties that can emerge from organizations of matter in the absence of any designer or other anticipator of those properties (http://serendip.brynmawr.edu/complexity/hth.html and http://serendip.brynmawr.edu/reflections/GrobsteinSoundings.doc).
Cashmore writes "Some will argue that free will could be explained by emergent properties that may be associated with neural networks. However, ... in the absence of any hint of a mechanism that affects the activities of atoms in a manner that is not a direct and unavoidable consequence of the forces of GES [genome, environment, stochasticism], this line of thinking is not informative in reference to the question of free will." Is Cashmore actually ignorant of the fact that one might have said the same thing a hundred and fifty years ago about inheritance? that most of the contemporary successes of biology have to do precisely with discovery of unexpected mechanisms by which larger scale organizations influence smaller scales ones, including the activities of atoms? Does Cashmore honestly think he (or any other set of biologists) is in possession of a catalogue of such possibilities adequate to preclude the existence of yet to be discovered mechanisms adequate to account for free will in some form?
Yes, stochasticity by itself does not yield free will in any meaningful sense. And yes, there are "causal" difficulties with conceptions of "free will" that conceive it either in a dualist/vitalist mode or as something not influenced by other nervous system activity. Both points are worth making (and have been in other contexts (cf Robert Kane's A Contemporary Introduction to Free Will) but neither requires rejecting the possibility of a nervous system organization that yields behaviors influenced by genes, by the environment, by stochasticity, AND by additional feature, a form of nervous system organization (itself a product of evolution) that yields some degree of personal choice as a consequence of both GES processes and its own influence on them. For some of my own recent thoughts in this direction, see http://serendip.brynmawr.edu/exchange/node/5643 and http://serendip.brynmawr.edu/exchange/node/3720 and links from there, as well as the older http://serendip.brynmawr.edu/bb/freewill1.html).
Cashmore, as a biologist, is fully entitled to make his own decisions about whether this is a possibility promising enough to devote his own time to it. What he is not entitled to do is to make such decisions for other biologists, nor to suggest to others outside the biological community that his is the only position that might be seriously taken by a biologist. To assert that an interest in pursuing the free will problem "is nothing less than a continuing belief in vitalism" is a serious misrepresentation of the motivation of biologists, like myself, who have an interest in the problem but none whatsoever in vitalism. Perhaps more importantly, it discourages work by others in a potentially productive area and suggests to people outside biology that biologists regard them as genuinely "conscious automata." Cashmore may be content to be one such, but I, as a biologist, would prefer to take a more agnostic and less confrontational position. In the interests of not only biological research and broader shared human understandings but also of maximizing the potential of humans, individually and collectively, to play a creative role in the universe of which we are a part.
The above notwithstanding, I fully agree with Cashmore that existing and anticipated understandings of the relation between the nervous system and human behavior will require revisions of a number of cultural practices, judicial procedures among them. And I'm inclined to agree with him that judicial proceedings should focus first on action, irrespective of any presumptions about the origins of such actions. Deciding what actions to discourage and encourage, in the interests of social organization, and doing so should be the overriding concern. Where Cashmore and I part company is that I would consider the existence of free will, and the potential for greater free will, in deciding how to deal with undesirable actions and to encourage desirable ones. Yes, some behavior doesn't involve free will but other behavior, on my account, does. Encouraging greater development of free will would, I suspect, do more to counter socially undesirable and promote socially desirable behavior than any procedure based on a presumption that we are all "automata." Am I sure of that? No, but it seems to me an experiment worth trying.
Finally, I think the Cashmore article is relevant to our discussion of science and "pseudo-science" (cf http://serendip.brynmawr.edu/exchange/evolsys/chance10#comment-119668 and following comments) with the latter understood not as "not science" but rather as "adopting the mannerisms of science in order to give greater weight to opinions held for reasons other than shared observations." The article seemed to me an exemplar of the latter. It was published in a prestigious scientific journal and appears to be a peer reviewed paper. What most people don't know is that this particular journal exercises serious peer review in the case of publications of authors who are not members of the National Academy. For authors, like Cashmore, who are, review is more or less pro forma. The article acquires further cachet by reference to the brain and neural activity but in fact contains at best passing description of actual observations. There are extensive references to distinguished scientific figures but these are primarily offered as "authoritative voices," rather than as sources of additional observations. Not good science by my criteria.
What's actually on Cashmore's mind, it seems to me, isn't the encouragement of continuing inquiry but rather an effort to disseminate a particular perspective about human life, that humans, flies, bacteria, and bowls of sugar are more or less the same thing. Indeed, in one sense they are, all are "bags of chemicals." But in other senses they are each quite different, and I find myself to be at least as interested in exploring the differences as in exploring the similarities. Its an interesting question why Cashmore is more inclined to the former perspective, and I to the latter one. I wonder to what extent Cashmore's perspective relates to his concern that my perspective "serves only to hinder what should be a major onslaught on determining the molecular genetic and chemical basis of behavior." Is he primarily concerned about his funding or is there something else that makes "free will" unappealing to him? Happy to entertain further inquiry into my own motivations, in the service of further discussion of the role of values/esthetics in science, and inquiry generally.
Anne Dalke, June 21, 2010 (7)
I want to pick up on the concluding paragraph of Paul's comment, asking "to what extent Cashmore's perspective" might be motivated by a concern about funding, or whether there might be "something else that makes 'free will' unappealing to him"--and, contrariwise, appealing to Paul. In other words, I'm interested in pursuing "further discussion of the role of values in science, and inquiry generally." What investments undergird the arguments each of us make? Where from do they arise? Are they modifiable?
What's feeding this particular stream of questions for me right now comes from a very unexpected place. On Labor Day weekend, my oldest daughter will be marrying into a Hindu family, and in preparation for both a very complex Quaker-Hindu ceremony and the life-long familial-and-cultural negotiations that will follow from it, I've been reading a book written by Swami Chinmayananda, and recommended to me by the father-in-law-to-be: called Self-Unfoldment, it offers an explication of Hindu beliefs. The central focus of the book is the need for rehabilitating our inner selves, as opposed to rearranging the external world. And what struck me instantly is the repeated use of the word "science" to characterize the understandings of the Hindu mystics, which have been acquired, over centuries, by observing human behavior and practicing its modification:
"I order to understand Nature...material scientists for centuries have been observing and studying the objective phenomena of the world. The ancient...sages or seers...pursuded a different system of study: They turned their attention to the subjective Self and studied the world around them from that subjective standpoint. This subjective study constitutes the Science of Reality...."
"Religion is the remedy for a particular unrest felt by the human being...the technique by which we get our minds and intellects trained to grasp and understand the larger themes of the universe and our own place in it. The science of the spirit has a very practical use for us....Religion...teaches us a method for creating in ourselves the equipoise to...meet...the ever-changing world...."
"Religion begins as a scientific reevaluation of life....While the material scientists take the outer world as their field of investigation, the subjective scientists take their own inner world of experiences as the field of their search for truth."
"Science is based on innumerable hypotheses that provide possible explanations of certain natural phenomena. As new data accumulate, the laws and theories of science may need revision. Thus, science is a growing tradition, the present research being performed on the basis of the truthfulness of past conclusions. The scriptures represent the data gathered and conclusions arrived at by generations of sages, the scientists of the spirit...."
Cashmore's work, Paul says, is "pseudoscience"; would you all grant to the work of these Hindu mystics the sobriquet "science"? Like Alice, I prefer language that is "subtle and nuanced, honest as to its reach and limits, and, preferably, to our own experiences and investments." By those lights, might self-conscious Hindu practice be called "scientific"? And might unconscious, unreflective "science"--science that doesn't (for example) acknowledge its presumptions and investments--be called "pseudo"? Might science that aims for the ideal of "objectivity," "non-personality" (or a standpoint that is "beyond personality") be neglecting the origin and motivation of inquiry in personal, located perspectives?