Moral Ecology

This is an extract from my Doctoral thesis, which was granted by the University of New South Wales. If you’re interested in reading the full version (warning: it’s 107,000 words), just click on the image below to be taken to a PDF.

Click here for a full PDF of the thesis.

Moral diversity is often dismissed as being something to explain away en route to discovering the correct answers to moral questions, particularly by those who adopt a realist metaethical perspective.

In this thesis I argue that moral diversity is actually far more interesting than this perspective might suggest, and that understanding the causes and dynamics of moral diversity can tell us something about the nature of morality itself.

I draw on the tools of evolutionary biology, game theory and moral psychology to give an account of the existence and variation of moral and norms, and how they change over time.

I argue that morality can be seen as a kind of cultural technology, the function of which is to help solve the problems of social living in order to facilitate prosocial and cooperative behaviour.

However, the optimal solutions to the problems of social living depend on the state of the physical and social environments in which those solutions operate.

As such, few, if any, norms or attitudes optimally satisfy the function of morality in every environment.

Furthermore, the existence of other norms or attitudes within an environment effectively alters that environment such as to impact the success of other norms and attitudes, introducing another element of dynamism to moral systems.

I call this phenomenon “moral ecology.”

I also argue that these dynamics have influenced the evolution of our moral psychology, introducing another source of diversity into moral attitudes.

This picture of morality suggests that moral diversity is not just an artefact of ignorance, error or bias but is a by-product of the inherent complexity of social living in a wide range of physical and social environments.

23 Comments

Mark Sloan · 25th November 2010 at 5:38 am

Tim, I like the idea that diversity in our moral biology is an adaptive trait. It occurred to me many years ago that a troop of chimpanzees with varying vigilance and aggression levels is likely to be, as a group, reproductively fit in more environments and circumstances than a troop with equal levels of vigilance and aggression. I had not carried that thought over to biology dependent moral perceptions and the (biological) psychological basis of conservatives and liberals.

However, this “diversity increases reproductive fitness” idea seems to me to run into problems if you try to apply it to cultural moral standards which I understand you to imply. Some cultural moral standards decrease reproductive fitness. There are a lot more selection forces for cultural moral standards than just human reproductive fitness. Moral behavior started coming untethered from reproductive fitness with the emergence of culture and, at least in advanced countries, is arguably largely now irrelevant to reproductive fitness.

I disagree with your statement there are no moral facts. I am almost certain there is at least one moral fact. That fact is something to the effect of “Moral behaviors are behaviors that increase, on average, the benefits of cooperation in groups by means of acts that are unselfish at least in the short term”.

Whether we are really disagreeing though is dependent on the kind of ‘ought’ we are thinking of when we say “moral fact”. The above claimed moral fact comes only with what I call a ‘rational choice ought’. That is, you ‘ought’ to accept the burdens of this claimed moral principle only if you expect it will better meet your needs and preferences (be your ‘rational choice’ in Rational Choice Theory terms) than your alternatives.

Another kind of ‘ought’ a moral fact might entail is what I call an ‘irrational choice ought’. That is, you ‘ought’ to accept the burdens of this claimed moral principle even if you expect it will NOT better meet your needs and preferences than your alternatives. I agree there are no moral facts that entail this kind of ‘ought’.

Note that an ‘irrational choice ought’ might be claimed to be justified by rational arguments such as Kant’s. The ‘irrational choice’ is in single quotes because it is in the terminology of Rational Choice Theory rather than philosophy.

Tim Dean · 25th November 2010 at 3:17 pm

Hi Mark. James made a similar point about reproductive fitness – I think I need to stress more clearly that morality evolved to promote fitness, but that’s not what it’s necessarily about today.

There’s a distinction between ‘adaptation’ and ‘adaptive’. The former is something that evolved because it lent an adaptive advantage in our evolutionary past. The latter is something that lends an adaptive advantage today. The appendix is an adaptation. Reading is adaptive.

I’d say morality is an adaptation, but that doesn’t mean it’s adaptive. It might be, but it’s not necessarily so. And there are aspects of morality that might be maladaptive today, such as our in-group/out-group categorisations. Likewise, our sweet tooth was an adaptation, but it’s maladaptive in today’s environment.

On moral facts: there are facts about morality, yes. But there are no normative facts; no facts that are intrinsically prescriptive. What you have stated is a descriptive fact about morality. It is very likely true. But accepting the truth of that fact doesn’t imply we ought to do anything. You’d need to add another premise that ‘we ought to increase cooperation’, or something similar. I don’t believe there are any objectively true prescriptive facts of this nature.

Instead, it’s up to us to agree what our fundamental oughts should be, then draw on descriptive facts about how best to achieve them. (Sam Harris is all about the second, and good on him – but he ignores/side-steps that first bit and just slaps down ‘well-being’ as the end of morality, like it’s incontestable.)

Not sure if I understand your ‘irrational choice ought’. You’re suggesting I should be moral even if it’s not in my best interests to be so? I think being moral requires sacrificing the option of free-riding, which probably will advance my interests more than being cooperative. But social contract theories suggest that it’s in my long-term interests to give up free-riding, and it’s in most people’s long-term interests to encourage me to give up free-riding. Is that kinda what you’re talking about?

Mark Sloan · 25th November 2010 at 5:12 pm

Tim, I can add the word descriptive (which is certainly true) to my claim about the existence of at least one moral fact. I might have said: …I am almost certain there is at least one DESCRIPTIVE moral fact. That fact is something to the effect of “Moral behaviors are behaviors that increase, on average, the benefits of cooperation in groups by means of acts that are unselfish at least in the short term”….

But I can defend the idea that in the case of this particular descriptive moral fact, there is no need for an additional assertion to the effect of “people ought (‘imperative ought’) to increase the benefits of cooperation” as is usually needed to produce a prescriptive statement. (An ‘imperative ought’ is an obligation that ‘should’ be accepted regardless of the individual’s needs and preferences.)

What makes this descriptive moral fact interesting is it appears that, if practiced as a moral principle, it would likely better meet the needs and preferences of people living in societies than any available secular alternative. Some people could accept those arguments and decide they ‘ought’ (‘rational choice ought’) to intellectually commit to the principle as their ‘rational choice’, the choice expected to best meet their needs and preferences. Under these circumstances, no assertions about imperative oughts are required.

No descriptive moral fact, including this one, inherently entails any oughts except the weakest kind, the ‘rational choice ought’. Fortunately for this moral fact’s potential future as part of a “workable secular moral system”, that may be sufficient.

When you say there are no moral facts, do you mean there are no moral facts either inherently entailing or otherwise accreting the force of ‘imperative oughts’? For that definition of moral facts, I fully agree.

But the fact that those kind of moral facts do not exist says nothing about whether a moral fact that entails a rational choice ought (the weakest ought) could exist.

Example of a ‘rational choice ought’. I prefer to not burn my hand on a hot stove. My stove is hot. Therefore, I ‘ought’ (‘rational choice ought’) to not touch the hot parts of my stove. No source of imperative force is needed.

Finally, if you are defining morality by the right moral principle (I suggest the principle based on what the function of moral behavior is in human societies) then yes, acting morally will almost always be expected to better meet your needs and preferences in the log term. Our predictions about the long term future in the moment decisions are made are often wrong. Our predictions about moral behaviors are in particular flawed because they often must be made almost instantly and moral behaviors are unselfish which our ‘base animal instincts’ motivate us to reject.

I look at it as “Should I rely on the wisdom of the ages or my own flawed perceptions and predictions is the heat of the moment about what action will actually best meet my needs and preferences over a lifetime?”

Tim Dean · 25th November 2010 at 6:35 pm

And again, we find ourselves in almost complete agreement.

I’d probably refer to your ‘rational choice ought’ as an instrumental or prudential ought (or a hypothetical imperative – even though they aren’t strictly ‘imperatives). And they’re the only kind I (we) reckon exist.

So everything start with an ‘if you want to x’, and the oughts follow based on how best (rationally) to achieve x.

Yet there are moral facts in the sense there are facts about how best to achieve x. Just not facts that say we bindingly ought to achieve x no matter what.

That’s anti-realism for ya!

Mark Sloan · 26th November 2010 at 6:43 am

Tim, I have gone back to read about the definitions of instrumental or prudential oughts and hypothetical imperatives to see if I can use them directly. It seems like a mess to me. It is so much easier to just define what I mean and move on!

Of course, I would like to use standard terminology, but I am going to have to study it more to be able to do that.

For example, I am not sure my ‘rational choice ought’ is the same as if “‘if you want to x’, and the oughts follow based on how best (rationally) to achieve x” because a ‘rational choice’ does not depend on our expectations, needs, or preferences being rational. Perhaps they are the same, but I like using the terminology of Rational Choice Theory because it specifies that expectations, needs, or preferences may be irrational.

Also, my understanding of hypothetical imperatives is that they are similarly focused on rational actions in a way that ‘rational choice’ oughts are not (I might be wrong about that).

In a related subject, I have written a paragraph that I thought might be of interest. It concerns how mere facts combined with ‘rational choices’ can provide useful moral guidance.

I prefer to meet my needs and preferences. I expect that committing to a moral principle that claims to be based on the function of almost all of cultural morality (the primary reason those cultural moralities exist in human societies) will better meet my needs and preferences than any alternative moral principle. Therefore, I ‘ought’ (‘rational choice ought’) to commit to that moral principle I expect will best meet my needs and preferences. By committing to it, I (may) accept that while cultural moral standards may be very useful in day-to-day life, they are only fallible heuristics for achieving the ‘ends’ of almost all of cultural moralities.

No source of imperative force (imperative oughts) is needed for these ‘mere facts’ to provide moral guidance.

Richard · 26th March 2011 at 10:07 pm

Comments from an interested amateur.

You have said what morality is not but not what it is. Is this an enterprise of discovery and if so how do you know it when you see it? For my money I would still describe morality as a truth seeking endeavour. More specifically I would define it as acting in pursuit of philosophical justice. A sense of morality is a measure of behaving justly, with regard to the consequences to all living things including regard to self but only to the extent that a just independent arbiter would; an ideal that does not necessarily preclude any of your views as far as I can see.

You subscribe to the view that morality evolved as survival benefit. I expect Darwinism has bearing on this. What I read is that morality evolved for the purpose of surviving or as an incidental observation that it assisted survival It may well have done. However, I see that morality may have alternatively evolved from empathy which then leads to Darwinian benefits. I think that is a significant distinction which does not contradict Darwinism. I may have misinterpreted.

I think I basically get what you are writing in Moral Psychology but I do not get what you are aiming at. If you are trying to say physiological predisposition determines morality then that is not explicit enough to come through to me. What I read is physiological and psychological impacts on morality.

Richard · 27th March 2011 at 9:12 am

‘If you are trying to say physiological predisposition …’ was meant to be ‘If you are trying to say psychological predisposition …’.

stonedead · 11th January 2012 at 1:53 pm

Your subject matter and approach is quite interesting!

I wonder if you have considered works of so-called “ecopsychology”? It examines how the environment affects the mind and behavior, and therefore how the condition of the “natural world” is bound up with the human condition.

Tim Dean · 11th January 2012 at 3:29 pm

Hi stonedead. I hadn’t heard of ecospychology before, but reading a bit on it now, it seems quite different to my broad approach. This might be because I use the term “environment” in a more narrow (or maybe more broad?) sense than in common parlance. Environment to me means the external circumstances relevant to some action or event. So the natural environment is one aspect, but so is the cultural and behavioural environment.

(BTW, I also use “natural” in a technical philosophical sense – meaning “of the natural world”, which basically includes everything, including humanity and the artefacts we’ve produced. So the opposite of “natural” is not “artificial”, as in the common definition, but “supernatural” or “non-natural”, meaning not governed by the laws of the natural universe.)

So moral ecology is saying moral norms adjust to the environment – including the cultural context and the actions and norms promoted by others in that culture.

Ecopsychology appears to want to bring people into closer connection with the natural (i.e. non-artificial) world. That’s a different programme, although one that, on the surface, I’m sympathetic towards.

GTChristie · 12th January 2014 at 11:30 pm

Tim I like the fact that reason plays a role in making moral judgments or decisions at some point in the process, even if it’s dead last (after impression and emotion have occurred). At least there’s room for reason there, where most of the 20th century left it all emotive. Reason is something an agent would need to engage on purpose, though. And many agents might never reason at all: going on the gut. Maybe that’s what marks the difference between village ethos and civilization: the latter depends on convincing people to stop and think before they hurt somebody.

Re: moral facts. The tentative fact Mark wrote out “…something to the effect of ‘Moral behaviors are behaviors that increase, on average, the benefits of cooperation in groups by means of acts that are unselfish at least in the short term’” contains too many cognates to be a naturally occurring fact of nature. It’s full of concepts and relationships among them that may or may not be facts themselves. Tim’s exposition is more of the form “moral behaviors evolved because” rather than “moral behaviors are” — and the “because” is fairly simple and direct. (After all, this is Ockham territory, right?)

Having said that (and always the skeptic), I think the “cooperation” vocabulary is useful (“good cooperators survived”) and that strategy is compatible with evolutionary theory (“certain emotions evolved”). But when I see “embarrassment” and “guilt” listed among the relevant emotions, I think we’re making things more elaborate than they really are. We have to be careful not to project our expected conclusions onto our premises. Embarrassment needs physically unique characteristics to distinguish it from “just another form of fear.” I get suspicious when definitions become too rareified. What has to be true, to call this a fact? is the question we need to ask repeatedly.

Mark Sloan · 13th January 2014 at 3:35 am

Hi GT,

I hope Tim does not mind my using his thread to reply to your objection to my proposed descriptive fact about morality as incompatible with it being a “naturally occurring fact”.

Its complexity is only superficial. It is due to my necessary use of existing language which was defined for other purposes.

I could equivalently state my claimed descriptive fact about what morality ‘is’ much more simply as “Moral behaviors are altruistic cooperation strategies”. But few people know what altruistic cooperation strategies are. (See Nowak “Evolution, Games, and God: The Principle of Cooperation” and Gintis “The Bounds of Reason: Game Theory and the Unification of the Social Sciences”.) The complexity you object to just defines altruistic cooperation strategies.

Alternatively, I could state my claim as “Moral behaviors are strategies that solve the cross species universal cooperation/exploitation dilemma”. This is the social dilemma of how to reliably obtain the benefits of cooperation, which commonly exposes one to exploitation, without being exploited – which would destroy future benefits of cooperation. Finding biological and cultural solutions to this universal dilemma is what made our ancestors such successful social animals.

I argue the ‘truth’ of my claim based on standard criteria for truth in science, but principally explanatory power for 1) the origins of the biology underlying our ‘moral’ emotions such as empathy, loyalty, shame, guilt, and indignation, 2) virtually all past and present enforced moral norms no matter how diverse, contradictory, and bizarre, and 3) Jonathon Haidt’s empirically discovered six cross culturally universal bases for making moral judgments: harm, fairness, freedom, loyalty, authority, and purity.

Also, including indignation, shame, and guilt as ‘moral emotions’ is not “projecting our expected conclusions”. Game theory shows that punishment is a necessary part of all altruistic cooperation strategies. This is why we commonly feel indignation when people violate moral norms we hold dear and think violators deserve punishment (of at least social disapproval). But punishing other people’s bad behavior is tricky and can backfire due to cycles of retribution. That is why our ancestors evolved shame and guilt, to efficiently provide internal punishment when we violate moral norms.

Biology makes no sense except in the light of evolution. Morality makes no sense either, except in the light of evolution.

Tim’s moral ecology seems to me to be consistent with my position. But Tim first adds the straight forward observation that, due to different circumstances, no single set of moral norms (altruistic cooperation strategies) will be able to maximize the benefits of cooperation in all societies and then he describes that observation’s implications for cultural moral codes.

Mark

GTChristie · 13th January 2014 at 12:54 pm

“Moral behaviors are altruistic cooperation strategies”
Well I know what that means. And it is succinct. And I admire it; it’s clever and better than the original statement. But in the succinct version “altruistic” is still an extra cognate there.

If this is a statement of empirical fact, then cooperation strategies that are not altruistic would falsify the definition.

To get out of that, you’d have to say that only altruistic cooperation strategies are moral behaviors. If you want to go with that, okay. But I’d have to disagree, if you go there.

The only point I was making: a complex process is going on, admittedly, but even the improved succinct description is preloading “cooperation” and “altruism” and “behavior is a strategy” into the definition of “moral behavior” before the process is understood. What if moral behavior is not actually about cooperation, but cooperation is just a happy by-product when everyone has the same moral framework? (Just to throw out an alternate possibility.)

I’m not criticizing your formulation (I tend to agree with what you’re after here). But you are stating a complex definition with too much in the bundle to verify as a single fact.

Mark Sloan · 13th January 2014 at 1:36 pm

GT,

Thanks for your reply.

However, I can readily defend the claim that “Moral behaviors are altruistic cooperation strategies” is factually ‘true’ in the normal provisional sense of science as a description of what morality ‘is’. However, “Moral behaviors are cooperation strategies” (which you are suggesting might be true?) includes behaviors, such as mutualism and coordination, which can increase the benefits of cooperation, but are not moral behaviors because they do not risk exploitation or require bearing some other cost. Risking exploitation in order to increase the benefits of cooperation defines the necessary ‘altruistic’ aspect of morality.

So far as I know, every single aspect of my claim about what morality ‘Is’ is required in order for the claim to be true in the normal sense of science. There are no extra parts. Leaving off any part, such as the altruistic aspect, would trash its explanatory power and lead to contradictions with known facts about past and present enforced moral codes and evolutionary game theory.

It all hangs together, or it all falls apart. I don’t see how to prove it ‘true’ element by element as you seem to be suggesting.

Mark

Tim Dean · 13th January 2014 at 4:06 pm

Interesting discussion. I’ll only add a few comments into the mix.

I’m sympathetic with Mark’s formulation of morality as “altruistic cooperation strategies”, but there are some problems to be overcome. First is moral norms that have existed that are either poor at promoting cooperation, or those that actually promote injustice or harm. The former might include norms that encourage such harsh punishments or strong retribution that they erode cooperation on average. The latter might be norms that entrench privilege in a minority at the expense of a majority.

To account for these I first distinguish between the normative form of moral norms and their content. On the first, a moral norm is any norm that appears to carry binding and inescapable authority (like a categorical imperative) – i.e. it can’t be overridden due to subjective preferences or prudential concerns. All cultures appear to have them, and they concern everything from proscriptions against harm to rules about dress codes and ways of speaking.

As for the content, I use functional definition, suggesting the function of morality has been to help solve the problems of social living and help facilitate prosocial and cooperative behaviour. Function here is in the biological sense – the function of an organ is that activity that explains why it was selected by evolution (i.e. the heart’s function is to pump blood, not to make a thumping noise).

However, few moral systems have been created with this function explicitly in mind, and have rather been innovated and spread according to cultural evolution – what has worked has tended to survive over time. Due to the messy cultural evolutionary process, we would expect there to be many sub-optimal moral norms – i.e. norms that are not terribly good at satisfying the function of morality. We would also expect there to be “corrupt” moral norms, which are created by those in power to entrench their power.

These sub-optimal and corrupt norms are moral only to the extent they possess the categorical normative form. However, in terms of their satisfying their function, they are not good moral norms, in the same way that a law enforced by a state that promotes injustice is still a law, it’s just not a good law.

As for moral emotions like guilt and empathy, these are the product of evolution, and do motivate moral behaviour. However, we have also evolved self-interested emotions, and the moral emotions are also error prone. That’s why we needed to invent normative systems enforced by punishment to elevate cooperation above the levels achievable through our limited built-in altruistic tendencies.

Don’t know if that clarifies any issues.

GTChristie · 14th January 2014 at 4:18 am

Although it’s primarily a discussion of Sam Harris’ “The Moral Landscape,” a major section in the middle of the following blog post discusses altruism and norms and some other related issues mentioned here. In the last paragraph of Tim’s reply above, the phrasing “we needed to invent normative systems … to elevate cooperation above the levels achievable through our limited built-in altruistic tendencies” reminded me that I’ve blogged about this in the past. There is a point where “moral emotions” (remember they are primitive) leave off, and norms (especially laws) begin. I have always said that a cultural process creates them (after all, that’s where the variation is, in human societies). Tim’s last points above point directly to a need to understand where biological evolution leave off, and cultural development begins. (I purposely avoid saying “cultural evolution,” mostly for clarity). Any “science” of morality needs to explain as much as possible biologically, but at the same time, attribute to biology only what can be selected by evolutionary process — and leave the rest to cultural processes. (I tend to see those as “cultural dynamics,” which explains my enthusiasm for Tim’s “moral dynamics” idea … if I’m brilliant on that point, so is he. LOL.