INTRODUCTION to CHAPTER THREE
This chapter deals with some prominent criticisms of the idea that Darwinian principles apply to social evolution. First we address the claim that Darwinism is inappropriate because it cannot deal with human intentionality. On the contrary, Darwin accepted that some species – including humans – were capable of reflecting on their circumstances, setting goals, and actin intentionally. But Darwin insisted that the evolution of intentionality and the development of intentions were caused, and subject to causal explanation.
Second we address the possibility of artificial selection. This is where some agent carries out a form of selection on a population. A misunderstanding here is to see it as a rival to selection more generally. Artificial selection is one form of selection. If it occurs in the social domain then that does not overturn Darwinism. On the contrary, Darwin used examples such as pigeon breeding to illustrate the more general idea of selection.
Third we address the phenomenon of self-organization, which is important in nature and in human society. Some authors have claimed that self-organization can serve as a general framework for understanding evolution. But self-organization addresses the emergence or development of a single entity, not a population of entities. Consequently it is an important but inadequate concept for dealing with evolution in complex population systems.
Fourth we address Ulrich Witt’s “continuity hypothesis”, which is a claim that nature grounds and constrains social evolution. We fully agree with this proposition. And it is entirely consistent with generalized Darwinian principles. So it would be a mistake to pose these two sets of ideas as mutually exclusive rivals.
KEY QUESTIONS
1. If some evolutionary biologists have downplayed deliberation or intentionality in their study of nature, does that mean that Darwinism necessarily excludes them?
2. How does the concept of artificial selection relate to selection per se? And how does the actual process of artificial selection relate to other processes of selection?
3. If self-organization is important, is it a sufficient organizing principle? If not, why not?
4. What is the relationship between the “continuity hypothesis” and generalized Darwinism?
Darwin's Conjecture 
H&K consider “intentionality as a human capacity that must have evolved from similar but less developed attributes among our prehuman ancestors” (page 48). Placed in this particular context intentionality must be object of selection, a process whereby intentionality will be created as well as annihilated, leaving the net result over the time-interval of selection as a residue of emergent purposefulness.
Here H&K touch upon an extremely important topic. I agree that intentionality is a feature of selection, but I deny that intentionality is object of selection. You can select ‘purposeful’, but you can’t select ‘purposefulness’. Thus we can regard entities as object of selection, but there is no intentionality embedded in what we select, although we might select it on purpose. If there were purposefulness embedded in what we select, what is the purposefulness of sigarets, a motorcar and an airplane? Is it the fact that sigarets are smoked, an airplane flies and a motorcar is for conventional transport on the ground? Well all three are the result of design and used for their own sake but it is difficult to see that there is a dynamic difference of internal purposefulness concealed in them. That there is selection between different items is a matter of the purpose attached to their selection, not a matter of different quantities of purposefulness assigned to their significance. Purpose is about why we select, not about what we select.
The general meaning of intentionality for stationary selection processes is in the subtle explanation offered by Claude Shannon and his ‘existence theorem’, the justification of which one may find in Shannon’s masterpiece [Shannon CE, 1948, “A Mathematical Theory of Communication”, Illinois Press, Urbana USA]. After it was published it was immediately clear that Shannon had hit on a theory of very wide general significance. I have extended Shannon’s argument to apply generally to all cases of ‘non-stationary’ (evolutionary) selection [Wallast, 2013]. Shannon’s explanation is based on the idea that we can map all potential variations (that can be selected during the time-interval of selection) uniquely onto binary prescripts: coded sequences with a length of k samples, provided k is chosen large enough (For historians and philosophers of science interested in the universality of this idea of mapping: Alan Turing used the same idea to demonstrate the limits of proof and computation, work that was closely connected with Kurt Gödel’s incompleteness theorems). The total number of different ‘potential variations’ that can be mapped in this way is (2 raised to the power k) as we may easily verify for sequences of samples that can only assume two different states (e.g. the states of population vs environment or the states of consumption sector vs investment sector etc.). Clearly, (2 to the power k) tends to infinity for k tending to infinity. This warrants that all potential prescripts/variations that can possibly be selected are each mapped one to one onto binary sequences which uniquely represent those potential variations. The following step consists of demonstrating that only variations are being selected from a much smaller subset of the set of (2 to the power k) potential variations. Hence selection is restricted to a subset of ‘typical variations’. It can be shown that there are no more than (2 raised to the power kH) typical variations each having equal chance (2 to the power –kH) to get selected. kH is the ‘Shannon entropy’ (the putty-clay of selection) that must be assigned to a typical variation in accordance with probability considerations. H is a positive number smaller than 1. From this it follows that the chance of selecting any typical variation equals the product of number and chance delivering (2 to the power kH) x (2 to the power –kH) = 1. That is, it is certain (i.e. an event with probability 1) that only typical variations get selected. It must consequently follow that virtually no potential variations different than typical variations get selected. Or more concretely: the chance of selecting a single ‘atypical variation’ is zero. We can only select typical prescripts/variations notwithstanding that there are infinitely many more potential prescripts/variations (in number: 2 to the power k) than typical variations (in number: 2 to the power kH) since k tends to infinity and 0< H < 1.
That atypical variations never get selected has the common interpretation that the selecting agencies do not select atypical junk (the variations/prescripts they never select) which is infinitely more available than the scarce number of typical variations. That agencies do select only typical variations has the common interpretation that they assign scarce value to the things they actually select whatever their further motivation to select. And clearly each single typical variation they do select is assigned the same weight/significance by these agencies. Thus we may just as well assign the predicate ‘purposeful’ to the selection of a typical variation by the selecting agencies in order to contrast it with the uselessness of selecting potential alternatives that actually never get selected: the ‘blind selection’ of atypical variations.
Why don’t agencies select atypical variations? Well apparently they select purposefully and the blind selection of junk is rather inappropriate within that context whereas the selection of typical variations is apparently only purposeful. This interpretation refutes the ideas of Jacques Monod, Dawkins and many other evolutionists that natural selection is random and blind. Well, it is random but it is not blind, but goal-directed. Evolutionary change is the fruit of intentionality, which is inherently linked up with what selection is: a goal-directed occupation of the selecting agencies involved.
Let us do some more surgery on this very general selection process (general because of the universality of the mapping operation which applies to all sorts of evolutionary systems). Note that the hypothetical blind selection of a sequence of k binary samples has an entropy equal to k in accordance with Boltzmann’s principle. On the other hand the entropy of a selected typical variation of length k is equal to kH, which is smaller than k. The question is: why don’t agencies select blindly as this seems to deliver more entropy? Well in fact agencies do select the blind sequence of entropy k only to rearrange it such that it is transformed into a new (more innovative) sequence of entropy kH smaller than k. In this particular context of the selection process the entropy k of the blind sequence is a measure of the uncertainty the selecting agencies face at the very start of selection. They select in order to reduce their uncertainty from k to (k – kH). The gain of ‘information’ they get by selection is equal to kH, the entropy of the typical variation. This implies that output selection must be considered as a process whereby (former) output k is rearranged to produce reordered output Y < k, resulting in a net reduction of output entropy equal to (k – Y). For input selection it is similar: a selection process whereby input k is processed into reordered input X < k, resulting in a net reduction of input entropy to (k – X). Thus the resulting net entropy gain of output over input selection is (k – X) – (k – Y) = (Y – X), which indeed is the surplus of output Y over input X as we might expect. Here (Y – X) is the net entropy growth of the system. Certainly the surplus (Y – X) does not contain a partial dynamic change of purposefulness over the time-interval of selection. The above explanation also clarifies why selection is not a process of optimizing some hazy object function (e.g. fitness). Only the uncertainties on output side and on input side of an evolving system are reduced by selection and this is the only way the selecting agencies collect information: i.e. just simply by entering into transaction events of exchange (entropy for entropy) and nothing else than that. These events may of course be caused, sustained and triggered by rational and even irrational considerations, by moral and amoral behavior, by good or bad luck, by detailed calculations of net profit or net cost, by financial capacity or lack of financial capacity, by manifest narcissism or by cautious reservedness, by natural disasters or whatever other occurrences, but all detailing discussion over such matters does not affect the conclusion that the selecting agencies are information-gatherers at output and input side by the very act of engaging themselves purposefully in events of exchange and that this is in itself sufficient to constitute the equations of growth and all the other equations that follow from it. The key to understanding evolution generally is in the Shannon-inspired mathematics of the selection processes of entropy and in his existence theorem.
In response to Len, Thorbjoern and I do NOT say that intentionality it is an object of selection. Rather, it is an OUTCOME of selection. We say that intentionality (= the capacity to have intentions) has evolved. Do you deny this, Len?
This may seem trivial, but creationists are not the only people to explicitly or implicitly deny that intentionality has evolved. Some scholars (such as George Shackle and James Buchanan) presumed that intentions were “uncaused causes”. This is inconsistent with a Darwinian evolutionary ontology.
Motor cars and airplanes do not have intentionality, according to our definition. Although we use the Shannon-Weaver definition of information, we do not use their definition of intentionality. Intentionality, for us, is the capacity to form goals by conscious prefiguration. (Some definitions of intentionality in philosophy are different from this.) There is no notion of consciousness in the Shannon-Weaver formulation, as far as I am aware.
Later in the book we make a distinction between “selection of” and “selection for”. (These terms are Elliott Sober’s but he uses them in a different way.) There is “selection of” objects, or entities. These entities are defined as interactors. Genes, for example, are not interactors, so the are not objects of selection. But the gene pool changes as interactors (entities) are selected, leading to selection for different gene frequencies. Because intentionality is a trait rather than an interactor, there is not SELECTION OF intentionality, and intentionality is not an object, but an outcome, of selection processes.
Len’s discussion (“surgery”) on general selection processes is best reserved for a later chapter, when we discuss selection.
LEN, YOUR COMMENTS ARE WELCOME AND INTERESTING, BUT PLEASE CONFINE THEM TO THE ISSUES RAISED IN THE CHAPTER.
Geoff and Thorbjörn, we come very close to the common understanding that intentionality is not an object of selection. It appears to me however that if intentionality is conceived as an outcome of selection, as you suggest, that this is another way of saying that it corresponds with the net output of selection, which is object of selection. I understand that you conceive consciousness as narrowly related to the presence of intentionality. I am inclined to consider consciousness as an object of selection, but I do not see the strict relation between intentionality and consciousness as you appear to suggest. Of course intentionality is always needed (or always present). The agencies involved happen to select in order to increase their contents of output and input information during the time-interval of selection. This results in a gross output gain Y of information at the output side and a separate gross input gain X of information at the input side. The net entropy gain is (Y – X) during the time-interval of selection. But we should be well aware that (Y – X) need not always be positive. It may be negative just as well resulting in economic recession and the decline of species and, if the negative trend is uninterrupted, to destruction and extinction. Well in the latter cases consciousness declines just as well if consciousness is object of selection. However there is still intentionality in selecting output variations of entropy Y and in selecting input variations of entropy X. And both Y and X may keep increasing even if (Y – X) remains negative. This example illustrates that intentionality does not condition consciousness or conversely. As you will notice, this conclusion is very much related to how we interpret the three premises of selection, variation and inheritance in GD. This all embracing basis makes it difficult always to follow strictly the chapter-order of subjects of your book. One example of a subject transcending the boundaries of chapter 3 is that you state in your reply that the interactors are the entities of selection. I do not deny that provided it is embedded in more elementary processes of selection occurring at a far smaller and much more fundamental micro scale of selection, whereby even smaller objects than replicators must be considered to be the object of selection. After all, the selection of k consecutive draws that constitute a variation does not rule out that, each time we draw a single micro-unit of selection, one sample of the k samples gets selected. Another example is your introduction of the concept of consciousness. A lot more can be said about that matter within the context of my interpretation of GD, involving questions like: Is selection (and the selection probabilities concerned) a subjective or an objective matter? Can we compare different evolutionary systems objectively and what about the real nature of the results of evolution that we observe all around us?
The bottom line is that intentions are real, and that (the capacity to produce) intentions must be explained. H&K suggest, “intentionality and other human mental capacities must have evolved from similar but less developed attributes among our prehuman ancestors” (p. 48). A contrasting viewpoint is that intentionality is a given, unique human property. We reject this view.
Even if lots of details remain in producing a complete explanation of the origins of intentionality, we suggest that such an explanation must be evolutionary.
In a later chapter (ch. 5) we discuss the nature of intentionality, but it is clear that intentionality introduces the possibility that humans (and other organisms) manipulate the environment in pursuit of their objectives. That is, we have the case of a nested selection process. The (intentional or non-intentional) manipulation of the environment is subject to higher order selection processes as well as a defining source of lower level selection processes.
Hello everybody. I’m struck by the excellent organisation of intellectual material lying behind this book and inspired by the courage of the authors in presenting a constructive research agenda (really hard, much more creative) rather than simply continuing with critique (less hard, can become a habit).
My comment is long and wide ranging, but it does fit Chapter 3. Chapter 3 deals with “… some prominent criticisms of the idea that Darwinian principles apply to social evolution”. I want to argue that Geoff and ThorbjØrn may have missed a reason for resistance to Generalised Darwinism (and thus missed another cause of criticism of the relevance of Darwinian principles in the social realm) amongst social scientists.
I think there’s another potential source of resistance to GD, from lay-persons and mediated through social scientists. I want to try to identify this source of resistance, and then offer a way around it which would be GD-compatible rather than use it as a way of seeking to undermine the GD approach.
1. I fear that social scientific resistance to GD may also arise from a belief by (some) social scientists that lay persons (who, after all, are in the majority, and are being ultimately theorised about in GD) will not ‘recognise themselves’ in the conception of humanity currently elaborated in the book as GD-compatible.
2. I also think that the same social scientists will not ‘recognise themselves’ in this conception of humanity.
3. I think it matters that those who create and use social scientific models of reality find the conception of the ‘agents’ in those models of reality recognisably human, are able to imagine themselves as the agents-in-the-model and are able to empathise with the predicament of the agents –in-the-model. It matters because without this empathy between model creator/user and agents-in-the-model, the use of such models eventually becomes a practice of ‘knowledge without feeling’. In the limit, model builders and model users lose a sense of caring about agents-in-the-model as fellow human beings, increasingly caring about something else rather more instead (theoretical consistency, being seen to be right etc.).
4. I am absolutely NOT asserting that GD ALREADY IS a practice of ‘knowledge without feeling’. I am asserting that there is something about the GD-compatible conceptual representation of human individuals in the book AT THE MOMENT which (in my view) significant numbers of lay persons and some social scientists will find does not accord with their understanding of ‘humanness’. I am suggesting that as long as this lack of accord persists, there will be social scientific resistance to GD and the practice of working within a GD framework risks becoming ‘knowledge without feeling’. Since I do not have unique access to what lay-persons or other social scientists ‘really’ think and feel, my claims may be wrong and I am open to being so persuaded, but here they are.
5. What is it about the general conceptual representation of human individuals in this book that is the cause of this lack of accord?
It isn’t the absence of ‘free will’ understood as the opposite of ‘determinism’ that is the cause of this lack of accord in my view. I am not saying that resistance is due to a need by social scientists and laypersons to cling to causal dualism, to recognise both subjective and objective causes as different and coexisting. I am sure some do resist GD for this reason, but that is not my argument here. My argument is independent of whether or not one accepts causal dualism or emergentist materialism. For my part I totally accept the GD critique of ‘mind first’ approaches to humanity as being unable to explain the origin of intentions and beliefs.
Instead, I think the GD-compatible representation of individuals this book adopts is (what I will call) a ‘cognition first’ approach to the brain-mind of individuals, and this is the problem. Such an approach, I claim, is unable to explain what causes the stability or security of each person’s cognitive structure or mental map and only able to explain instability or insecurity by appeal to some kind of genetic defect – a problem with biological instinctual development.
To explain further: My understanding of the conceptual representation of individuals in the book is one in which instincts (biologically evolved urges, reflexes and primitive emotions) are the platform on which (socially/institutionally evolved) habitual actions and habitual thoughts/beliefs are generated. From the platform of habits, ideas and more developed emotions emerge. Central to this conception of persons is an instinct-habit nexus or the individual’s internal cognitive structure or ‘mental map’ or thinking system. Emotional development is regarded as secondary: in part a very long run biologically evolved ‘primitive’ (e.g. ‘fight/flight’) capacity and in part a derived outcome of the operation of the instinct habit nexus or the internal cognitive structure. Given the focus is the social realm, very long run emotional development via biological selection processes can usually be ignored: primitive emotional development or capacity can be taken as ‘given’. Since the mental map or cognitive structure or the instinct-habit nexus generates other emotional development (i.e. the way individuals think about things affects how they feel), any remaining consideration of emotional development then becomes a second-order concern. The cognitive system, the instinct-habit nexus, is primary. Hence the ‘cognition first’ label. Of course I may be wrong in my interpretation here.
Suppose though I am roughly right in my portrayal of the GD-compatible individual. Then there is a tacit assumption in the GD-compatible individual that the mental map or instinct habit nexus of any given person has an innate stability or secureness. Put another way, each person is assumed to be ‘secure’ about their ‘sense of self’, secure about their way of thinking about themselves and the world. It is known that not everyone in reality is so secure (in fact it is likely that many people are not – and indeed one might explain much entrepreneurial and other high-achieving socio-economically relevant activity by this fact). The only explanation of self-insecurity or lack of cognitive security consistent with this view of the individual is one rooted in genetic defects, affecting the biological development of the individual’s ‘primitive’ emotions classified under ‘instincts’. This is oddly similar to the rationalist conception of individuals or ‘Standard Social Science Model’ (SSSM). The GD-compatible view is ‘better’ in that the mechanism of thinking (instinct-habit nexus) is more explicit and multi-layered, but like the SSSM it simply assumes (without explaining the causes of) a psychological robustness.
The GD-compatible view of the individual’s mind seems to have little conception of the role of emotional development (lay persons call it ‘wisdom’, or ‘self knowledge’) as NECESSARY for self security, necessary for the stability of the operation and expansion of the instinct-habit nexus or internal cognitive system or mental map of each person. There seems to be no recognition of emotional development as a capacity or set of capacities that enables the instinct-habit nexus to operate – that enables cognitive capabilities to function and grow. Nor is there then any room for recognising the SOCIAL roots of emotional capacity development in the quantity, quality and variety of social relationships individuals have access to.
To reiterate, since I don’t have superior access to what lay-persons really do feel or think I could well be wrong, but I would now argue (i) – (v):
(i) I think lay persons tacitly understand that human persons are not fundamentally cognitive but are fundamentally emotional beings.
(ii)I think (i) is because lay persons at some level are aware that emotional development is necessary for cognitive stability or the security of a person’s sense of self.
(iii) I also think lay persons continue to warm to the fluffy, vague and overused word ‘community’ because at some intuitive level in their own life experience (or lack of it) they see that emotional capacity development occurs through the quantity, quality and variety of interpersonal relationships individuals are embedded in.
(iv) I think that lay persons will sense that this understanding both of the central role of emotional capacity in sustaining cognitive capability (i.e. the instinct-habit nexus) and the role of community in supporting emotional capacity is missing in the present GD-consistent conceptualisation of individuals.
(v) I think that this sense of a missing dimension critical to ‘humanness’ will contribute to resistance to GD from both lay persons and social scientists.
6. Given there is a problem, what would be a way forward?
Instead of the present view of biologically evolved instincts (urges, reflexes, primitive emotions) being the platform for habits and thence ideas and more developed emotions, I propose this alternative. Biologically evolved instincts (as now, including primitive emotions) are the platform for emotional capacities (that develop out of the quantity, variety and quality of social relationships in which persons are embedded). Emotional capacities and instincts are then the platform for the development, exercise and growth of habits and thence to ideas and other emotions as before.
Essentially, this is an ‘emotion-first’ (rather than ‘cognition-first’) evolutionary view of the mind of human persons. I am claiming an ‘emotion-first’ view is more likely to be in line with the perception of what it is to be human from lay persons and which many social scientists would also support.
7. In our roles as ’scientists’ and ‘academics’ we tend to be governed by two predominant habitual beliefs, I fear. Our ‘scientist’ role deals with the question of ‘humanness’ by asking what is it that is special about humans as distinct from other species? We then use this as a criterion for conceptualising humanness. So scientists tend to operate on the habitual belief that cognitive powers are central to being human and emphasise these. Our ‘academic’ role encourages the habitual belief that the emotions are secondary to and regulated by the intellect. However, lay-persons who are neither scientists nor academics may alert us to how we can get caught in our habitual beliefs. Lay-persons may be more willing to attribute ‘humanness’ to qualities of particular people they interact with – people who are particularly ‘together’ or wise relative to others. Lay persons may be less wedded to the intellect, perhaps more aware of the place of self knowledge, of wisdom in checking the excesses of the intellect. If the biological and the social are to be unified in an overarching framework successfully, perhaps we need to pay closer attention to the majority of those who are neither scientists nor academics but are the people we theorise about. It is the duty of social scientists, in particular, to do this. I think if we do so, we are more likely to build theoretical representations of the world (and in particular of ourselves) with which both theorists and lay-persons can empathise. Approaches consistent with GD can then be practices where we feel our knowledge. Or so I wish to suggest.
Thanks for sticking with this, if you have come this far!
Thanks especially to Geoff and ThorbjØrn for providing this opportunity to comment. Sorry it was so long.
As to our intentionality discussion, I have had some further reflections about the matter and I think now that I have been too rash in asserting that purposefulness is not object of selection. Shannon’s existence theorem motivates that the selecting agencies select information to reduce uncertainty. Information is here the semantic equivalent of meaning (as reflecting the subjective ideas of the selecting agencies) and, in turn, meaning is much like purposefulness. It is therefore perhaps correct to conclude that purposefulness is object of selection. In this regard intentionality is reflected in the information gathered though it is literally not a complete equivalent of purposefulness. It reflects more or less the subjective aim of the selecting agencies to select purposeful. However, all of it is a matter of definitions. I agree with Geoff and Thorbjörn that intentionality is not a unique human property and I also agree now that intentionality (as far as it reflects purposefulness ) evolves in all sorts of evolutionary systems.
Len’s rethink is much appreciated. We now agree that intentionality is a selected trait. (I shall leave until later our distinction between outcomes (selection for) and objects (selection of) of selection. We also agree that some intentionality may exist outside humans. But it is important also to emphasize that fully-fledged conscious deliberation, involving concepts and language, is unique to homo sapiens.
I would like to start by posing a counterquestion to the first question for Chapter 3: If some evolutionary economist downplayed a natural science account to intentionality, does that mean that Darwinism necessarily excludes it? I wish to argue that behaviorology, a natural science that emerged from biology, would not only exclude intentionality, but other folk psychology terms (some of which appear in H & K, e.g., habits). In place of such psychological terms, behaviorology offers materialistic, selectionistic, agencyless explanations of behavior, established by over seven decades of experimental research—both basic and applied—along with systematic conceptual work. We may also call this discipline Skinnerian science, but because this label has diverse meanings—as does the more common designation, behavior analysis—behaviorology is a more precise name, agencylessness being its defining feature. In this context, agency is any account of behavior that presumes an inner hypothetical entity (e.g., “The cat ran away because it was scared,” “the student has difficulty reading because he is dyslexia,” or habit, which I shall address later on).
I have found that within the community institutional/evolutionary economists there are opposing views of B. F. Skinner (1904–1990). Some believe him to be a behaviorist like J. B. Watson (1878–1958), importing mechanistic scientism into the study of human behavior; that is, stimulus—response mechanical causation. Whereas, other institutional/evolutionary economists assert that the institutional approach takes the position of Skinnerian behaviorism, grounding the roots of human action in institutional structures instead of in individual preferences. Here is the bottom line: To quote Skinner, “A stimulus-response formula has no answer, but operant behavior is the very field of purpose and intention” (About Behaviorism, 1974, p. 55).
We must sharply distinguish between Skinner’s operant conditioning and Pavlov’s S—R conditioning (also called respondent or classical conditioning), but at the same time note that both types of conditioning are the result of natural selection, and they may occur concomitantly within an organism. In both cases, certain mechanisms evolved as an outcome of enabling individuals—during a lifetime—to acquire behavior appropriate to the demands of novel environments. An illustration of respondent condition would be the experienced runner’s cardiac speedup just before the start of a race (a stimulus—response relation). With operant conditioning, behavior is selected by its consequences; that is, a behavior is emitted, an event immediately follows (the consequence), and subsequently in the same or similar situation that behavior becomes more likely to occur (i.e., the behavior has been reinforced). To illustrate, you are working on your computer and accidentally press a key combination that instantly clears your screen of all the open windows, thus giving you access to files that were hidden. When you press the key combination again, the open windows return. You continue to repeat that routine in the future. We could now say that your action of pressing the appropriate key combination was selected by its consequence— the open documents promptly disappear or reappear. In other words, the action was reinforced. (Note: lay language, respondent and operant conditioning are called involuntary and voluntary behavior, respectively. But that is not a useful differentiation.)
Another source of confusion is between phylogenic and ontogenetic contingencies, in that both appear to be associated with intention or purpose. True, they both shape and maintain behavior—but in totally different ways. Skinner (1969) writes, “We say that spiders spin webs in order to catch flies and that men set nets in order to catch fish. The ‘order’ is temporal” (p. 192). A complete account must elucidate upon the effects. Flies or fish not caught cannot affect behavior; only past effects can. Spiders building the more effective webs are more likely to be have offspring that survive; the man who set his net so that it catches fish will likely repeat that action—his movements have been reinforced.
There are many other issues entailing the distinction between psychology and behaviorology that I would like to bring before evolutionary economists, but that will have to come with another chapter—probably Chapter 5, since it deals with selection. What I have not touched upon, but of key importance in relating behaviorology to evolutionary economics is cultural selection. This consideration will come to terms with the argument that Skinner’s theory is moribund by methodological individualism. I will end for now by recommending a reading:
Skinner, B. F. (1981). Selection by consequences. Science, 213, 501–504).
Downloadable: http://www2.psychology.uiowa.edu/classes//31174/documents/selection%20by%20consequences.pdf
Chris Fuller’s comment is interesting, but also very long, so I shall address a couple of points only.
First, he suggests that there may be additional reasons for resistance to generalized Darwinism (GD). If I understand him correctly, the resisance would stem from a view that GD does not portray humans as understood, or as they really are. In response, GD cannot be ADEQUATE to portray humans, because human populations have features that are not shared with other complex population systems, such as mice or robots. The fact that GD is GENERALmeans that it cannot include attributes that are found among humans, but not mice or robots. Consequently, we need additional principles or theories to deal with humans that are not covered by GD. That is why GD is “not enough”, as we say in the book.
Second, Chris raises the hoary old question of free will and causal determination. In philosophy the “principle of determinacy” (POD) is the notion that “every event has a cause”. So, according to the POD, intentions must have been caused (and intentionality as a capacity must have evolved). We inbterpret Darwin as upholding the POD.
So is the POD compatible with free will? Philosophers are divided on this question, and it depends how free will is defined. The majority (known as “compatibilists”) answer in the affirmative, saying that a meaningful notion of free will is consistemt with the POD. (See the entries on “Compatibilism” in Wikipedia and in the online Stanford Encyclopedia of Philosophy, but beware of ill-defined usages of the multi-meaninged word “determinism”.) Personally I think that much of what goes under “free will” – including real choice and personal responsibility for actions – is consistent with the POD. Obversely, notions such as an “uncaused cause” are incomptible with a Darwinian evolutionary perspective.
Chris rightly puts his finger on a point of possible resistance here among social scientists. Many accounts of “agency” in social science dance around the questions without staring them in the face. The ambigious notion of “determinism” is treated as a cardinal sin. Much of that stuff is junk. One of the good things about GD is that it raises the question of our connection with our evolved legacy, while retaining notions of human uniqueness.
Thanks for your reply Geoff. In response:
[1] OK, I accept the point that GD is a framework not a theory in itself and will need more specific theories to make clear causal connections. My worry in a nutshell was that the general GD framework as portrayed in the book doesn’t over-tightly specify the conception of the mind (i.e. over-emphasise cognition – this is what I meant by a ‘cognition first’ view) so leading to the inability to explain what causes stability in human cognitive structures.
[2] I tried to avoid the free will- determinism issue because my main point [1] above didn’t depend on it. I’ve frankly really struggled with that issue over the years. I was always convinced by the importance of habit and the influence of institutions in shaping the mind via habitual development ( so I accepted a strong form of ‘downward causation’ by institutions on individuals -what Geoff has called ‘reconstitutive downwards causation’). A sophisticated determinist position as in GD is the most comfortable place to sit if you adopt this. But then I became queasy about determinism. I took on board social theorist Margaret Archer’s concern that (essentially) determinism implies that in principle experts can know our own minds better than we can (implications for ‘social control’ and all that…) and so (to avoid that) I adopted a causal dualist position. Causal dualism is most comfortable with a ‘mind first’ view: the minds of individuals (their ‘internal conversations’ driven by innate ‘concerns’) can be causes distinct from objective causes emanating from objective structures. And yet…I agree with Geoff that the mind first view is untenable as an explanation: it shuts out a way of explaining where these ‘concerns’ come from. Its taken me a long time but recently I’ve become more persuaded of a compatibilist position (the idea that rather than free will being in opposition to determinism, some notion of free will can be consistent with determinism).
Chris, I now mostly agree with you. But the ambiguous weasel word “determinism” still causes some havoc. You write of “Margaret Archer’s concern that (essentially) determinism implies that in principle experts can know our own minds better than we can”. But when defining the principle of determinacy (= every event has a cause) it is important to point out that this is an ontological rather than an epistemological doctrine. Saying “every event has a cause” does not imply that “experts” can “know” or predict anything. It is important not to conflate the principle of determinacy with other very different notions such as predictability determinism (being able to predict future events) or regularity determinism (if event X then event Y). The multiple-meaninged word “determinism” must always be defined more clearly.
(See my 2004 book and my 2004 Journal of Economic Methodology article on all this. Details are on my personal website)
Geoff, thanks for the help here. I was aware of the different definitions you cite but your reply makes clear to me that the worry about ‘social control by experts’ I had is not implied by the sort of determinism GD adheres to.
Hi all, catching up again. Yet another lively discussion! So if GD is not adequate in portraying humans and their complex behaviour, and if this is a general theory, then should we not be thinking about what other auxillary theories or explanations are required? Maybe this could be envisaged like a cladogram (used in evolutionary biology to describe the relationships between species ie ancestor – descendent). So GD could be the ancestor at the top of the tree, with branches lower down describing the aspects of the social world that these principles cannot explain? Just a thought.
Enjoying all the debate around intentionality, and of course this is the result of evolution. Being consciously aware of your environment, of dangers, predators etc as well as food sources and safety would have been critical to our ancestor’s survival. Being able to remember all of this, to consider, to think, and to communicate with others would clearly have a selective advantage. Any individual with these characteristics would be likely to survive and pass this on. Clearly this is rooted in our ancestral past. Maybe on of the auxillary theories we need is around intentionality of thought and action in the social world. How our ideas become tangible through communication and agreement? If intentionality is important than perhaps self organizing systems do feature somewhere in all of this. Is thought a self – organizing system? Is society?
Now onto one or two of your questions. I think that artificial selection is directly related to natural selection. Most simply because Darwin himself was a breeder of fancy pigeons so he had first hand knowledge of artificial selection didn’t he? So it stands to reason he would look for the same actions in nature, and thus come up with the principle of natural selection. There is a relationship between these two concepts. Artificial selection is what we humans do, and perhaps this is a good principle to use when trying to understand why people or groups act the way they do, or think the way they do. Maybe it is a better principle to use than say natural selection? But you may disagree. I think it is difficult to say for sure whether human actions are purely down to natural attributes or to artificial ones. However society and the social world are surely artificial constructs?
Not sure I agree that nature grounds and constrains social evolution. Societies have been evolving for millennia as our ideas and awareness develop. There is no sign of this grounding to a halt. What are the limits or constraints and can these be defined at all? What is the evidence to suggest this?
“Not sure I agree that nature grounds and constrains social evolution. Societies have been evolving for millennia as our ideas and awareness develop. There is no sign of this grounding to a halt. What are the limits or constraints and can these be defined at all? What is the evidence to suggest this?“
One constraint is that all human societies must deal with the biological problems of maintaining a population of Homo sapiens. Even something as simple as the human need for food, water, and shelter have constrained constrained social evolution along certain paths, have resulted in similar patterns of social evolution across the world, and shall continue to limit social evolution in the future.
In addition, human beings are a species that reproduced individualistically. This mammalian characteristic is not a necessary feature of an eusocial species and its presence will help determine the constitution of the possible array of societies that human action can maintain. A contrasting example is the collective reproduction of ants, and how ant-societies are limited in their own constitution by the need to cope with the problems of their form of reproduction.
I want to respond to Melissa’s and Harrison’s misgivings over Ulrich Witt’s “continuity hypothesis”. Both of you rightly point out that natural and biological constraints can be shifted (at least to some extent); and human society can sometimes shift them. Hence there is causal interaction between the social and the natural. Nothing here contradicts the continuity hypothesis as presented by Witt. One issue, of course, is how much natural constraints can be shifted. But in general terms the continuity hypothesis accepts socially-shifted constraints. In fact, in the view of Thorbjoern and myself, the continuity hypothesis is true but rather trivial.
Yes, I accept that there are natural constraints such as limitations on the amount of food that can be grown to feed a human population, or that there is a finite amount of water on the planet, and that consumption of certain fuels could be going to have a harmful effect (possibly already is) on human populations. But what I mean’t was constraints on societal ideas, such as to do with technologies, morals, awareness, behaviours, construction of institutions and agreed norms etc. Evolution of social awareness. I am not sure there are limits to this. To our ideas on how to evolve.
Just caught up now on the thread and wanted to pick up on a specific pointy in chpt 3. There is reference to misunderstandings and meaning around the term blind (p49) and the example is used from Dawkins “We are survival machines – robot vehicles blindly programmed to preserve the selfish molecules known as genes”. The suggestion here is that the term ‘blind’ misleads in its connotations with blindness in terms of “a depiction of organisms including humans with little conception of what they are doing” (p48).
The point I have been thinking over is that Dawkins did not write “robot vehicles programmed to blindly preserve…” but instead wrote “robot vehicles blindly programmed…”. It is then the programmer who is blind not the robot vehicles and as such the statement still allows for the consciousness of the vehicles. It is the act of programming which Dawkins is suggesting where there is not conscious ‘guiding hand’.
Enspace is right to underline the important point that social or natural order is (mostly) undesigned and that no-one designed our genes. But a problem I have is with Dawkins’s incautious use of the metaphor that humans are “robots”. This suggests a possible lack of capacities for consciousness, moral reflection, aesthetic appreciation, and so on. Dawkins’ account misleads people into thinking that Darwinism implies that humans are unconscious, unreflective, calculating machines. People picked this up as soon as the Selfish Gene was published.
Did Dawkins mean this when he wrote The Selfish Gene? I don’t know. I believe it is important to recognise consciousness and intentionality in human action and interaction. How much these interactions are driven by the gene is a good question. It is perhaps an over simplification to see us solely in terms of a vehicle for reproduction only. As for robots……..not sure that human ideas or customs can be reduced to behavioural programme like concepts.
If we are evolutionary thinkers then we have to accept that there is no guiding hand at work. This is a process which is governed by the system itself.
I take Geoffrey’s point that its not just Dawkins’ use of the term ‘blindly’ but also his decision to liken our existence to mechanical beings which gives rise to our interpretation that he is seeing human consciousness as pretty irrelevant from an evolutionary point of view.
Human consciousness is essential I think in understanding evolution in social systems.
Melissa (May 20 @ 12-19, 2nd para) said “Is thought a self organising system..?”
Most of you have probably heard of the ‘lateral thinking’ guru Edward de Bono. A few years ago I tried to look at the ideas behind his lateral thinking concept of the late 1960s. De Bono claimed the notion of lateral thinking wasn’t a gimmick but derived from a theory of the operation of the mind he developed as, yes, a self-organising system. His obscure and unusually organised book The Mechanism of Mind (1969) spells this theory out (he had a disdain for academia despite originally being an academic researching the interaction of biological systems in the body at Cambridge for his PhD so the book avoided giving scholarly references!) I see it as an attempt at an explicit theory of how the mind forms habitual thought processes and found it stimulating, though it needs work to follow. Perhaps someone might want to investigate..? I’m fond of de Bono’s explanation of the linkage between humour and creativity: much better than the dry approach of Steven Pinker for instance. Perhaps then T. Veblen’s Darwinian emphasis on the habitual nature of mind could be developed into the idea that the essence to being human is having a sense of humour, because both humour and creativity come from the same habitual mind processes. I am being serious here.
Strangely enough I have been reading Edward de Bono recently and yes he does describe the brain as such a system. Have you read Luhmann? He is quite interesting, as his theories are around self-replicating and self-reproducing systems. Such systems produce their own elements ie thoughts and communications which are situated in a specific space and time. In his theory of autopoiesis he states that social systems include disorder as well as order and that reproduction occurs through integration and disintegration. Which is a curious thought isn’t it? So to him maintaining a social system is not just about replication of the same patterns, ie maybe the idea of a meme suggests this, but that the production of different elements or entities that are different from the previous ones is important. So thoughts produce more thoughts, possibly organisations produce more organisations, but these differ from one another in some way
A key component of society being communication. How is this transmitted in society? Is it replicated exactly or does it change/evolve through interaction with other entities? If replication is exact then this reduces variation over time, so a consideration is what is the relationship between replication and variation? Is it not important that there are errors or mistakes, or slight changes in the shape of ideas that are communicated in society? Is this perhaps a factor in change?