Evolution and the argument against moral realism
In this paper I defend Sharon Street’s arguments in her paper ‘A Darwinian Dilemma for Realist Theories of Value’ (Street, 2006) in order to argue that the most tenable explanation for evaluative judgements makes no reference to independent evaluative truths. However, I go further by exploring the possibility of evolutionary theory extending a blanket scepticism across not just moral realism, but realism relating to all supposed independently existent truths (aesthetic, scientific, mathematic et cetera). I explore this as a form of transcendental argument. The contingent ‘fitness’ of beliefs, as well as the instruments upon which these beliefs are formed (the human brain), are the necessary conditions behind having any beliefs at all. Subsequently, if the independent beliefs we practically assume to be true have formed through their instrumental role in aiding fitness, this may suggest that, if there is objective non-evolutionarily-conditioned objective true-beliefs, we can have no access to them.
Introduction
In this essay I will be agreeing with the statement. Evolutionary theory can explain why we have the moral beliefs we do and this does tell against moral realism. I will be defending Sharon Street’s arguments in her paper ‘A Darwinian Dilemma for Realist Theories of Value’ (Street, 2006) in order to argue that the most tenable explanation for evaluative judgements makes no reference to independent evaluative truths. However, I wish to go further by exploring the possibility of evolutionary theory extending a blanket scepticism across not just moral realism, but realism relating to all supposed independently existent truths (aesthetic, scientific, mathematic et cetera). I agree that evolutionary theory positively tells against moral realism. However, I am interested in exploring whether this is not, in fact, a criticism that can be made against all forms of realism. There may, therefore, not be a special threat to moral realism that is not also faced by myriad other realist belief systems. I wish to suggest that, if our evaluative judgements are a result of the contingencies of environmental fitness, then this charge may actually extend to all beliefs. This is because we are evolved creatures, and our instruments of knowledge, including our perceptive faculties, our reason et cetera, are evolved also. The threat for realism is that evolution appears to demonstrate that our beliefs are targeted first and foremost at utility rather than truth. I wish to explore this as a kind of transcendental argument. The contingent ‘fitness’ of beliefs, as well as the instruments upon which these beliefs are formed (the human brain), are the necessary conditions behind having any beliefs at all. Subsequently, if the independent beliefs we practically assume to be true have formed through their instrumental role in aiding fitness, this may suggest that, if there is objective non-evolutionarily-conditioned objective true-beliefs, we can have no access to them. Therefore, the ‘Darwinian Threat’ (Street, 2006), has the potential to be problematic for not only the moral realist. In some sense then, this deflects fire from the moral realist in so far as his is only one form of realism amongst many threatened by evolutionary theory. Having said this, the full scope of sceptical thinking introduced by evolutionary considerations, cannot be adequately covered or resolved in this essay. I wish, nonetheless, to explore it and to suggest some areas for further research.
In §1 I will layout Street’s arguments against moral realism. In §2 I will consider realist responses to Street’s position as well as the possible reply from anti-moral realists that reason itself is evolutionarily conditioned and therefore affords us no privileged access to non-evolutionarily conditioned ways of thinking. Finally, in §3 I will briefly consider whether this form of epistemic scepticism is at all tenable and is not, in fact, self-refuting for the sceptic.
§1 The Darwinian Dilemma and arguments against moral realism
Moral realists contend that moral facts exist. They are true independent of any given individuals’ beliefs. They propose that there is factual content held in these statements. The statements are cognitive in the sense that they are correctly (or incorrectly) making assertions about an independently existing state of affairs. Killing is objectively right or wrong in a way analogous to scientific assertions; water molecules objectively do have a fixed and concrete molecular structure independent of any (unlikely) disagreement. Moral realism is certainly attractive. To know that our moral intuitions do correctly map an independently existing reality has the ability to create a secure basis for a system of moral thought, that is, if they can be accessed. And indeed this has been the pursuit of moral philosophers since Plato.
Evolutionary theory presents a unique threat to realist theories of value because it demonstrates the social efficacy and thus highly environment-dependent nature of moral assertions. Moral beliefs have a strong basis in a contingent evolutionary history. The argument, in a highly simplified form, states:
​
Premise 1: There is selective pressure on the content and nature of our moral judgements/intuitions etc.
Premise 2: Natural selection primarily favours beliefs that aid survival, not necessarily beliefs that are true (even if true beliefs may often aid survival).
Sub-Conclusion: Our moral judgements/intuitions have evolved and serve/have served a purpose in a contingent evolutionary history.
Conclusion: It is therefore far from clear how our moral judgements/intuitions can be said to be correctly identifying independently existing necessary truths.
It is claimed then, by Street and others, that morality is best understood as an adaptive illusion (Street, 2006). An illusion because our evaluative judgements do not correctly map onto some real independently existing reality and adaptive because it is extremely beneficial from a Darwinian perspective to hold these beliefs. Of the almost infinite range of evaluative judgements it would be possible to hold, the beliefs most people do hold are extremely adaptive; beliefs such as ‘it is wrong to eat your children’ or ‘the fact that one has treated one well is reason to treat that person well in return’ (Street, 2006, p115). The individual who held the opposite views would not survive for long (nor would the genes that encouraged these sentiments). Street admits that there is a complication here, in so far as the content of evaluative judgements are not inheritable. However, Street claims, it is perhaps the basic non-evaluative judgements which are passed on. For example, emotional responses of disgust at certain behaviours or fear of pain which are then socially and linguistically translated into ‘this behaviour is bad’ (Street, 2006, pp118-119). Thus, human moral sentiments are selected for pain-avoidance, survival, propagation of genes and co-operation and are contingent means to an end rather than ‘real’ in any cognitive sense.
For Street, there are two potential options available to the realist in their reply. Firstly, they could deny a relation between evolved evaluative judgements and independent evaluative truths, thus allowing moral truths the luxury of independence from evolutionary pressures. The difficulty here is that many of the same evolved evaluative judgements are those that the realist wishes to state are objectively true. However, without a relation between the two, the odds of our evaluative judgements correctly mapping onto these truths would be staggeringly low. The vicissitudes of evolution, the physical, environmental and social pressures sculpting our evaluative judgements, would have to somehow be ‘pushed’ in the direction of an independent objective truth that they bare no relation to. Alternatively, it could be a chance coincidence that our evolved sentiments align with independent moral truth. If this is the case, however, then it is more than likely that we are off track with our evaluative judgements because of the sheer range of possible judgements (Street, 2006, p122).
Secondly, the realist may like to claim that rational reflection identifies independent moral truth and that this stands, to a greater or lesser extent, independent of outside evolutionary influences. This is a position that I will be returning to later in the essay. Street dismisses this position because she believes that the only possible manner of rationally assessing evaluative judgements is in terms of their consistency with other evaluative judgements which are themselves, evolutionarily conditioned (Street, 2006, p124). For Street, the most plausible route for the realist is to assert a relation between evaluative judgements conditioned by evolutionary forces and independent evaluative truths. However, the only possible relation for the realist is a ‘tracking relation’ (Street, 2006, p125). This is the view that somehow evaluative judgements have evolved to correctly ‘map-onto’ or correlate with objectively existent evaluative truths. The argument states that we have been selected for our ability to correctly access independent moral truth. Street dismisses this as standing very poorly alongside other scientific hypotheses such as the ‘adaptive link account’ (Street, 2006, p127). This account simply maintains that evaluative judgements have, non-cognitively, been adapted to allow for survival and not necessarily for access to objective truth. Street believes that, in many cases, there is no advantage to our evaluative judgements tracking the truth, especially as in some cases it may be disadvantageous. This can be seen in other areas of cognition such as perception. For example, it is disadvantageous to ‘see’ electromagnetic rays rays despite their being objectively existent because a perception of them has no selective advantage for humans and is therefore a distraction. Therefore, correspondingly and expectedly, no sense faculty has evolved in order to recognise them (Street, 2006, p130). The tracking account also does not explain how we have evolved many useful evaluative judgements which we later believe to be wrong. For example the distrust and inhumane treatment of group outsiders (Street, 2006, p133). It also has less parsimony, by which she means it is inefficient to evolve faculties which track moral truth rather than evolve faculties which simply create an instinctive emotive response to ill-adaptive behaviours.
I agree with Street in her favouring the adaptive model over the tracking model, not only for reasons suggested by Street but also because of its ability to pass the often illusive tests for adaptiveness. Firstly, there is cross-species correlation. The closer the genetic relative, the more we can recognise our own ‘human’ moral sentiments and the deleterious effects on those individuals who do not adhere to them; for example, a sense of fairness and justice in capuchin monkeys. It is difficult to see what the realist can make of this. Most realists would be reluctant to hold the position that non-human animals are able to access independent evaluative truths. Therefore, given the physical and evolutionary similarities between species, an evolved capacity to adhere to group norms appears much more likely. Secondly, game theory as an independent calculation of optimums suggests that the moral sentiments most of us hold do likely lead to greater benefits in communities. Dawkins demonstrates in ‘The Selfish Gene’ how co-operative strategies develop in order to maximise outcomes rather than as a tracker of independent truth (Dawkins, 2016, p261).
§2 Possible defences of moral realism and issues with these
Critics of Street may counter with accusations of the genetic fallacy. Here the charge is that it would be fallacious to throw doubt on our moral beliefs merely because of their dubious epistemic origins. The question becomes, to what extent does an adaptive model of the origins of evaluative judgements suggest that a belief’s content is itself merely contingent? The scope of this attack is, however, limited. Street herself pre-empts it in her paper. If the moral realist wishes to maintain that no relationship exists between evaluative attitudes and independent evaluative truths then it would seem extraordinarily unlikely that we have correctly ‘mapped onto’ those truths with no evolutionary pressure to do so. It would be, to use Street’s metaphor, like expecting a boat with no sails to be blown from the UK to Bermuda (Street, 2006, p121). Therefore, the origins of the belief really do throw doubt on the belief’s veracity. Another possible response, suggested by Russ Shafer-Landau, is that realists may wish to deny that there are as many possible evaluative judgements as Street suggests (Shafer-Landau, 2012). Here the suggestion is that the realist doesn’t need to provide explanations for how evolution has allowed our evaluative judgements to correctly ‘map onto’ objective moral truth. This is because there are only certain possible evaluative judgements that may semantically pass for ‘morality’ in the first place (Shafer-Landau, 2012, p11). To use the same analogy, much less explanation is needed to account for the boat being blown from the UK to France than is needed to account for the boat being blown from the UK to Bermuda. Shafer-Landau states that the pre-conditions of morality are that we take only certain statements as moral ones. Presumably, ‘eat your children’, would never pass for morality in any conceivable evolved state of affairs and thus, the range of possible evaluative statements the realist needs to account for evolution to have ‘missed’ is not as wide. I do not find Shafer-Landau’s response convincing, largely because we do not know what could pass as morality in a different evolutionary history. Human evolutionary history provides us with a sample group of one. As Darwin expresses in ‘The Descent of Man’(Darwin, 2017), morality would look very different if we had the evolutionary history of the bee or the morality of hamsters. Therefore, we simply cannot say that what we consider to be the semantic parameters of morality are necessary for a definition of morality in some a priori way.
Another realist approach may be to respond by arguing that human reason, as a faculty of mind, stands outside the environment of selective pressure. Therefore, while not denying that a great deal of our evaluative attitudes are evolved responses, we have also evolved the ability independently to assess these moral truths from the neutral stance of speculative reason. The issue I take with this defence is that the realist provides no compelling reason to maintain that rationality is exempt from evolutionary pressure and, therefore, introduces a circularity into his argument: reason can access non-evolutionarily conditioned moral truths while also being conditioned by evolution. Shafer-Landau appears to admit this, stating ‘we might ask whether such reasoning will really be immune to the influence of evolutionary pressures?’ (Shafer-Landau, 2012, p14). Therefore, there is no way of independently assessing how accurate our reason is for tracking any supposed independent truth rather than simply a contingent evolutionarily advantageous false belief. This seems to be what Street is saying in her response to value naturalists: ‘I judge that Hitler was morally depraved...someone being morally depraved is roughly identical to having such and such a character...But on relying on these and other evaluative judgements, I rely on judgements that are saturated with evolutionary influence’ (Street, 2006, pp139-140). There is no reason to view this sceptical concern as not extending to non-moral as well as moral beliefs.
Arguably, in the realist’s defence, it could be maintained that pure speculative abstract reason of the sort that may access independent truth (moral, aesthetic, mathematical or physical) is a spandrel of practical evolutionarily advantageous reason. Therefore, we are able to access truths which are not just contingent on survival and fitness because evolution has given us a free gift of speculative reason. The argument here is that pure speculative reason is not itself adaptive, instead it is a benign bi-product of highly adaptive practical reason. This allows us to think about and access objective truths. William Fitzpatrick makes a similar point in his article ‘Debunking Evolutionary Debunking of Ethical Realism’(Fitzpatrick 2015, p886). His view is that natural selection may account for cognitive capacities but they then become non-evolutionarily developed in cultural contexts and training. Other examples of this include writing. We have not been selected for an ability to write, which is a recent historical phenomenon evolutionarily speaking, but it is a benign bi-product of our ability to speak which is highly adaptive.
However, I do not believe that Fitzpatrick’s response is successfull because the ability to refine reason must surely be seen as an adaptation. It certainly is optimal in any community as it affords great advantage in communities of humans, for example, in gaining deeper conceptual understandings and thus manipulation of an environment. Street also considers the possibility that certain cognitive faculties may be a bi-product of our evolutionary history. However, she states that, if there is no relation between the evolution of a faculty and its bi-product, then it is extraordinarily coincidental that a bi-product of an evolutionary function is able to successfully track truth (Street, 2006, pp142-143). Street claims that the realist might respond by arguing that traits arriving as an entirely accidental bi-product of another adaptive trait happens all the time in evolution. This doesn’t work for Street, however, because the trait out of which the fluke developed is so complex and unnecessary. We simply do not need a complex faculty that independently tracks moral truths, however we do need a trait which tracks moral instrumentality. Writing isn’t much more complex than speaking and thus may be considered a bi-product. However, a faculty that accesses independent truth is much more complex than a faculty that creates instrumental usefulness in communities. There is, therefore, an issue of parsimony. It appears inefficient to have developed a faculty which accesses independent moral truth. Furthermore, the supposed faculty coincidentally also tracks the same moral truths that we have evolved to believe in.
§3 Is scepticism tenable?
Having explained why I believe moral realism is seriously undermined by evolutionary theory, I now wish to consider whether these same arguments cannot be extended to cover all forms of realism. If our cognitive faculties have developed to aid survival as their primary aim rather than objective ‘truth’, then this may also throw doubt on the tracking ability of all our beliefs. For example, perhaps water is not truly H20 and 2+2≠4, it is simply evolutionarily advantageous to believe so. A criticism of this position would be that evolutionary theory tells against moral realism more than the realism of non-moral beliefs. For example, evolution doesn’t require that our moral beliefs and attitudes track moral truth. Morality is concerned with interactions and relationships with others, not necessarily with a physical universe outside of our interactions. It is therefore evolutionarily advantageous to believe that certain interpersonal norms are objective and transcend human beings when they do not because they then become more categorical and therefore binding for that community. A similar argument can be made when examining religious belief: it may be advantageous to believe in a deity regardless of a deities existence. However, it is not evolutionarily advantageous to believe in the objective validity of physical laws when they are not true. Therefore, evolution does require that our beliefs about our physical universe do track physical truth. If not we would be falling off cliffs, believing we could fly, et cetera.
The anti-realist may wish to respond by arguing that you cannot really know the independent existence of physical laws because the way that we come to know them is through reason, which is itself an evolved instrumental faculty that aids fitness selection. Our knowledge of all the terms involved, cliff, fall, gravity, are all contingently based on perception and reason which are evolved faculties not necessarily revealing the full picture of the objective world. This can be seen with the earlier example, we do not ‘see’ electromagnetic rays. Therefore, there is a veil of perception/cognition that we cannot penetrate: evolutionary fitness adaptation. This line of thought arrives at a Kantian conclusion: that our manner of cognizing the world; geometry, space, time, aesthetics, morality (all modes of thought) are, as far as we are aware, transcendental categories through which we must filter experience. Evolutionary theory demonstrates that our faculties have evolved to be necessary categories through which we must structure our world. They are necessary because it is hard to see how one would survive without them. However, the sceptical turn comes when we consider that selection promotes beliefs that aid survival not truth. If truth aids survival, this is good, but evolutionary theory appears to demonstrate that truth is only good instrumentally in so far as it aids survival; it is not the ultimate aim. Therefore, perhaps, in Kantian language, we can only know the evolutionarily phenomenal, and not the non-evolutionarily-conditioned noumenal. There is no way of thinking outside of the categories we have evolved to think within. We therefore cannot know the realism of anything. It is not certain whether this is a tenable position, however, it certainly has the possibility of casting serious doubts upon our beliefs
As interesting as this position is, I believe that it is ultimately unattractive. Firstly, the introduction of evolutionary theory does not cover any new ground in the epistemic debate that idealist projects such as Berkeley’s have not already covered. That there may be a distinction between our ideas and reality isn’t a new idea in philosophy; evolutionary theory simply gives us another reason to suppose that there is or may be. Secondly, global scepticism leaves us intellectually impotent because there is no way out of the solipsism introduced by scepticism once it has been introduced. The introduction of global scepticism is therefore a dead-end in terms of further investigation. Finally, if all modes of thought are evolutionarily conditioned and therefore circumspect, how can a debunking argument get off the ground in the first place? Extending sceptical doubts to reason is then self-refuting and viscously circular. Shaffer-Landau warns of this: ‘Of course, it is possible that selective pressures have thoroughly saturated our moral faculties in doxastically discriminating ways to the point that any exercise in moral thinking is bound to produce tainted moral beliefs. But this possibility alone does not justify skepticism’ (Shaffer-Landau, 2012, p20).
Conclusion
I have aimed to show that evolutionary theory does tell against moral realism. I have argued broadly in favour of Street’s arguments, in particular that there is much greater parsimony in the anti-realist than the moral realist account. I have defended Street’s position by looking at the selective adaptive advantage of instrumental moral beliefs. I have also explored counter arguments to the realist’s position, particularly from Shaffer-Landau. I have also looked at the possibility of extending the anti-realist claims to include all beliefs. There is not the scope in this essay for coming to a satisfactory conclusion on this issue. My own embryonic view is that the sceptical argument is a compelling one but that it also leaves the sceptic in much the same position as any radical epistemic sceptic. Alvin Plantinga makes a similar argument (Plantinga, 1993, p159). So also does Shaffer-Landau, who states ‘A wholesale scepticism is implausible, and is in any event self-determined for debunkers, since they can achieve their aims only by advancing philosophical arguments’ (Shaffer-Landau, 2012, p125). Further work in this area would extend outside of evolutionary theory and draw on more diverse sources within epistemology and the philosophy of science.
Bibliography
Darwin, C, ‘The Descent of Man and Selection in Relation to Sex’, (2017), Hardpress
Dawkins, R, ‘The Selfish Gene’, (2016), Oxford University Press
FitzPatrick, W, J, ‘Debunking evolutionary debunking of ethical realism’, (2015), ‘Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition’, Vol. 172, No. 4, S
Joyce, R, ‘Irrealism and the Genealogy of Morals’, (2013), John Wiley & Sons Ltd Ratio
(new series) XXVI 4
Plantinga, A, ‘Warrant and Proper Function’, (1993), Oxford University Press
Shafer-Landau, R, ‘Evolutionary Debunking, Moral Realism and Moral Knowledge’, (2012), Journal of Ethics and Social Philosophy, Vol. 7, No.1
Sober, E, ‘Philosophy of Biology’, (2000), Westview Press
Street, S, ‘A Darwinian Dilemma for Realist Theories of Value’, (2006), Philosophical Studies