|Third Sunday Roundtable|
Join our list to get occasional updates about meetings and important events. You can be removed from the list at any time without hassle.
Past Debates: Elliot Sober and David Sloan Wilson
Online chat with Elliot Sober and David Sloan Wilson (dsw) on the evolution and psychology of altruism
5 September 2001
Massimo: Elliott, David, welcome and thanks so much for agreeing to spend the evening with us. The discussion will loosely focus on the themes of 'Unto Others', that is on group selection, altruism, selfish genes, psychology, philosophy and then some more. OK guys, anybody wants to start?
ester: I haven't read your book. Could you give in a nutshell definition of altruism pls?
jholmes3: Is there a theory out there on altruism and group selection. I mean, I had the impression that according to Darwin, selection acts on individuals, not groups.
elliott: there are two meanings for the term altruism: psychological altruism has to do with the motives individuals have; evolutionary altruism has to do with the fitness consequences of a behavior. A mindless individual can be an evolutionary altruist, but can't be a psych altruist.
Massimo: What's your response to the standard anti-group selection argument, that groups reproduce at a much slower rate than individuals, and therefore group selection is bound to be overrun by individual selection?
elliott: Darwin invented the idea of group selection although he gave individual selection explanations 99% of the time.
jholmes3: so in other words, an entire group may be selected for if individuals within that groups behave altruistically...
elliott: there can be group selection even when groups are "mixed," containing both selfish and altruistic individuals.
dsw In answer to Massimo, this argument is based on a conception of groups as multigenerational units. Many kinds of groups are of a shorter duration, for example many social groups. That's a quick answer and there are other reasons why even multigenerational groups can be potent units of selection.
elliott: the central idea is that groups must differ in how many altruists they contain groups with more altruists are more productive than groups with fewer altruists.
jholmes3: ok, that makes sense
<Joyce>: Do you see individuals as either altruistic or selfish? Or each falling along a continuum?
Massimo: But, David (and believe me, I'm actually pretty sympathetic with your position), don't social groups of ephemeral duration pose a different problem, namely that of the cohesion of the group itself to count as a "unit"?
elliott: when Joyce asks whether we think that individuals are alt or not, i take it she means psychological altruism correct?
elliott: ok. our view is that people sometimes have both altruistic and selfish motives both. We don't claim that people are exclusively motivated by selfish motives or that they are exclusively altruistic.
Wed Sep 05 20:18:28 EDT 2001: dsw : dsw In answer to Joyce, thinking of discrete altruistic and selfish types is just a convenience. In most cases there will be a continuum. There is interesting evidence for this in humans that I may try to describe later.
<Joyce>: Got it!
ester: it's not the behavior as such that is altruistic or not, it depends on the motive, right? What about secondary motives?
elliott: that's how psych altruism is defined, not evolutionary alt. Motivation has nothing to do with the definition of evolutionary altruism.
dsw Responding to Massimo: The question "what is a group" is obviously fundamental to the whole subject. It's hard to do it justice in such a short space but here is an attempt. In everyday language, a group is the set of people with whom one interacts with respect to a given activity. This is a discussion group; it includes the people who are talking and no one else. My family is another group, my class still another, my nation still another and so on. There is an infinite variety of groups but only because there is an infinite variety of activities. For any particular activity, there is a single appropriate group. When we think about the evolution of social behavior, we are usually trying to understand a given behavior: alarm calls, cooperative hunting, and so on. The behavior defines the appropriate group for animal social behavior, just as human social behavior.
scott: How small can a group be? A group of genes?
elliott: groups can contain just 2 individuals -- the classic setting of game theory.
Massimo: Scott's comments about genes brings up the parallel idea that even genes are much less 'stable' over time than people think. For example because of recombination in sexually reproducing organisms.
elliott: multi level selection theory involves the following parallelism: individuals are to groups of individuals, as genes are to individuals. There is a part/whole relation, and the opportunity for parts to cooperate or be in conflict with respect to the welfare of the whole.
dsw In an effort to merge with the other comments, an individual is indeed a group of genes that influence each other's fitness, and therefore a very appropriate unit of selection. The fact that individuals are ephemeral is irrelevant as far as group selection theory is concerned. The question is, does the gene spread at the expense of other genes within the group, or does it spread by causing its group to do well, relative to other groups?
Massimo: OK, about your take on psychology (since we have a psychologist among us). Daniel Batson has accused you of a partial disservice because you claim in chapter 8 of Unto Others that the psychological evidence is inconclusive. Specifically, he claims that the "open-set problem" (the fact that egoists can always come up with modifications of their theory to fit the experiments) is no ground to suspend judgment. In essence, he thinks you are being too generous to the egoist camp. Are you?
scott: So, what is whole is relational to parts. And wholes can be parts of higher-level wholes?
Massimo: Scott, I think that's exactly it.
elliott: in reply to Massimo, I still disagree with Batson's criticism. He thinks he has solved the Open set problem, showing that no version of egoism (at least none that anyone has formulated explains the behavior. We disagree.
ester: what are the factors playing a role in the development of psych. altruism?
elliott: I'm afraid we didn't study this developmental question but just concentrated on the problem of determining whether adults in groups within groups ever have altruistic ultimate motives. If we are right that they do, the question remains of how they developed that trait. Human social behavior represents a good example of many nested groups.
Massimo: But Elliott, while reading Unto Others I often got the impression that egoism becomes increasingly less falsifiable and more ludicrous in order to meet the objections raised against it.
elliott: it is true that we bent over backwards to give egoism every break. We didn't want to be accused of refuting egoism by not exploring the full resources of the view.
Massimo: very altruistic of you... ;-)
elliott: making psychological egoism an easy target for refutation has been something that lots of philosophers have done before.
dsw In answer to Massimo: That's a problem with the existing literature! We took it seriously enough to pursue egoism arguments to the bitter end. If we exposed the ludicrous nature of the arguments, that is a good day's work!
<Joyce>: What is the argument for egoism?
elliott: reply to Joyce: here's one -- everything people do can be explained by some selfish motive or other. so there is no need to invoke altruistic motives. to do so is unparsimonious
ester: yea, it gets even trickier when subconscious psychodynamic motives are taken into account.
Massimo: Speaking of egoism. This group often discusses Ayn Rand in relation to egoism/altruism. What's your guys' take - if any - on her 'philosophy'?
elliott: I don't think much of it. She defends egoism as a normative theory, as well as a descriptive one. She thinks we ought to be selfish because the attempt to be altruistic (sometimes, often, always???) leads to bad results. She also thinks that the only folks that matter are the great and gifted. the others are like bugs and their lives have no value.
<Joyce>: to Elliot:Now, we're back to unselfish...what if one "feels better" if they are sacrificing for the group?
ester: sounds like Nietzsche's views.
dsw I actually discuss Ayn Rand in a paper on the many meanings of altruism and selfishness in everyday language. Rand seems to celebrate selfishness but her concept of objectivism often involves extreme altruism. For example, if you and I are competing for a job and you are better qualified, as an objectivist I should want to you have the job.
elliott: yes, there is a resemblance between Nietzsche and Rand.
Massimo: ok, we still got Joyce's last question hanging.
Carl As I recall Rand defined altrusim as sacrificing a higher value to a lower one. I am not sure the words are used by her as you use them.
elliott: ok. on Joyce's question -- feeling good is a selfish goal. so if that's the only reason we help, then egoism has the right explanation of helping
ester: Carl could you give a concrete example pls?
elliott: i recall that Rand defines altruism so that it would be irrational to be altruistic it would involve sacrificing something higher for something lower...
jholmes3: But if altruism is better for the survival of the group, then perhaps "feeling good" is a side effect of altruism that evolved to promote altruistic behavior.
elliott: jholmes3 is pointing to a possible evolutionary explanation of psychological egoism (the fact that we feel good when we help, and the desire to feel good motivates us to help).
dsw In general, there are at least two common meanings of the word "selfish" and associated words. One meaning defines selfishness at "Increasing your short term welfare at the expense of others and maybe even at the expense of your own long-germ welfare." By this meaning, statements "you'll be sorry if you're selfish" and "you'll be better off if you're altruistic to others" make sense. The other common meaning of the word selfish is "whatever increases the long-term welfare, including behaviors that are highly other-oriented." By this meaning, the statement "you'll be sorry if you're selfish" becomes a contradiction of terms.
Carl it depends on what one places value on but by her definition it seems that there is a value placed on making rational choices which can include helping the group.
elliott: I think Rand regards the desire to help the group as confused and irrational and weak.
Massimo: Still thinking about Joyce's question. Doesn't the position that if you help others to feel good you are an egoist beg the question of why do we feel good while helping others? BTW, obviously chats lend themselves to nonlinear conversations... ;-)
elliott: if feeling good is just a side effect of helping, then we may be altruistically motivated, but if feeling good is the only reason we help, then we're egoists.
Massimo: Elliott, right, but *why* do we feel good while helping?
<Joyce>: But isn't the individual "safer" when the group is better off? Could that be why they feel good?
elliott: maybe evolution has built us so that we feel good when we help, or so an egoist might propose
Carl Since feelings and thinking are integrally bound (see Damasioís book), don't the distinctions between them become meaningless?
elliott: if we help the group just to increase our own safety, we're egoists. but if we care about he group for reasons that go beyond concern for self benefit, we're altruistically motivated.
Andy I think a concrete definition of the egoist-altruist distinction requires a quantification of short vs long range considerations. An extension of economic utility theory, perhaps.
elliott: I think that egoists can be both short term and long term in their calculations. An egoist could take his or her total lifetime as the temporal unit during which self benefit is to be maximized.
Phil It seems we are suggesting either altruism OR egoism. Why can't a theory embrace both (to some degree)?
Andy Many apparently self-compromising decisions can pay a huge interest in the long term.
elliott: in UO, we endorse "motivational pluralism," which is the idea that people have altruistic AND selfish ultimate motives.
dsw Perhaps it is time to introduce the distinction between proximate and ultimate that is so important in evolution theory. Natural selection only "sees" phenotypic traits such as behaviors. Group selection favors traits that cause groups to perform better than other groups, regardless of the underlying mechanisms. In general, groups function well only because their members provide benefits for each other. But other-oriented behaviors can be caused by many mechanisms, only some of which count as psychologically altruistic.
Massimo: Of course, as you also point out at the beginning of UO, sometimes it's better to have an egoist do something that results in common good than a well-intentioned altruist messing things up...
dsw Right, in fact several people argue that conventional altruism is bad for groups, especially large modern groups. The nobel laureate Hayak made this point. How do you or Hayak define unconventional altruism?
Andy The selective pressure of a psychology compatible with efficient waging of war must have had an inordinate effect on any problems caused by "altruism".
dsw Continuing the same thread, many aspects of human society and probably animal society compell individuals to behave in the interests of the group. It isn't left up to voluntary self-sacrifice. But such systems still evolved by group selection because they work well compared to less organized groups.
elliott: there are two ways to benefit one's group at the expense of other groups -- help the individuals in your group and hurt the individuals in other groups. Group selection can lead both types of traits to evolve.
<Joyce>: I have spent much of my life, sometimes neglecting my own children whom I love dearly, to help children that are the victims of abuse. I was not abused as a child, but my thoughts and feelings about their suffering motivate me to sacrifice my own comfort to help them. Am I doing this because I am altruistic or because I can't stand my own pain? Therefore, an egoist?
elliott: this is hard to say, and I don't trust introspection to reveal to us what our real motives are, but ask yourself this: if you could take a pill that would make you stop empathizing with kids who are victims of abuse, would you? If not, this suggests that your motives in helping aren't just to make yourself feel better.
Massimo: Joyce, another question would be: does it *really* matter?
dsw To Ester: As I recall, Hayak thought that unfettered private enterprise made societies work at a large scale. The social control that underpins altruism and restrains selfishness at a small scale becomes stultifying at a large scale, especially in the form of centralized governments.
<Joyce>: to Massimo: No. I wrote an essay in seventh grade titled "Selfishness is the root of all good"!
Massimo: Joyce, I wouldn't go quite *that* far... ;-)
Massimo: Elliott, David, I see a movement by liberal thinkers such as yourselves and Peter Singer to take back evolutionary theory from the grip of Social Darwinism and claim that biology and morality can make for a good mix. While I personally agree with your take, weren't most liberals also harshly criticizing the same move when it was made by conservatives to serve and justify their own class and economic interests?
Barbara I never admired people like Mother Teresa.
elliott: I don't think that science should be done in the service of ideology. not that this doesn't happen, mind you. It is a mistake for people on the left to think that a biological explanation is necessary socially reactionary.
Massimo: right, but I'm sure Herbert Spencer was also thinking of not doing science in the service of ideology...Is that difference that we probably got it right this time?
elliott: Let everyone judge what the evidence is. i don't think that whatever political motives David and I have are particularly relevant. the question is whether the evidence supports claims of group selection, or not...
scott: The human genome project is ideological.
Sharron: Why, Scott?
scott: Why? To serve the human medical needs, not something else, say, ecology or other.
Massimo: Scott, but that goes for all applied science, doesn't it?
dsw I think it can be effectively argued that evolutionary biology offers an even playing field with respect to ideological issues. The important thing is guard the facts and attack faulty inferences. The fact that Hitler used evolution to justify his policy doesn't make evolution wrong. The inferences were wrong.
scott: Massimo: yes, all "applied" science is flavored with ideology. Since humans do science, humans bias it. If lions could talk...and do science. ;) what would we observe?
jholmes3: what would THEY observe!
Massimo: David, Elliott, thanks so much for being with us tonight! I'm sure everybody enjoyed the opportunity, and we look forward to see Elliott on campus for Darwin Day next February.
elliott: I've enjoyed the chat time (which was my REAL motive for participating). See you folks in TN on Darwin day.