It wasn’t my intention to take on the issue of evolution. I decided to tackle it in the discussion of AI in AgainstAgency, thinking an analogy in biology would explain why conferring agency on AI is a mistake, like it is in evolution.
I think it is better not to consider evolution as a ‘thing’ but rather as a theory, or at least a way of looking at change that make sense of a number of things that are counterintuitive.
Like in the Old Man River song, the river is indifferent to the troubles the people suffer, and evolution as a process is unconnected to any of the concerns people face. It just keeps rolling along in a process unconnected to any that people understand.
Change happens. People are dying every day. They’re dying like flies, considered at the time scales at which the human genome changes and evolution works, quite longer than the months required for fly genome change.
Death of individuals in the short term of a natural life is a regular thing, so it’s understandable. Other non-biological things also wear out.
But homo sapiens could also die out in the long term, because the resources people need to sustain life are not guaranteed to exist in the long term even if their supply seems regular in the short term.
This is imaginable, but homo sapiens seems determined not to die out in the short term. As individuals die, their grandchildren or great grandchildren are born, replacing them in greater or lesser numbers. Reproduction leads to preservation. It seems like a goal of homo sapiens is not to die out in the long term, but to live potentially forever through the power of sexual reproduction.
It is a truth widely, if not universally, accepted that evolution has no goals. Evolutionary change just keeps rolling along. Which seems to be incompatible with the goal of homo sapiens not to die out.
Saying change in biology has goals is like explaining the world as being the result of God’s handiwork. That explanation shuts down understanding rather than advances it. Goals are important elsewhere, and end-means analysis is important in explaining how to reach them, but goals, and by implication, agency, doesn’t help understand biological change or even devise ways to prevent species extinction.
Consider the accepted idea that evolutionary change is not goal oriented. Individuals are goal-oriented. Preservation of species by the procreation of individuals of the species to which they belong appears goal-oriented but the proliferation of certain individuals and disappearance of others by selection is not an end, either success or failure, to which evolutionary change is working.
Life needs the constant supply of resources, which is not guaranteed to exist. And when it ends, life ends, too. Certain groups, if not all groups, die out.
This is not something that anything or any group is working toward, unless they don’t like the group. For its part, evolutionary change just keeps rolling along.
Nevertheless, there persists the mistaken idea that evolution is a process of improvement guided by trial and error.
The idea is that wrong paths can be backed out of, or at least deprecated, and the right paths (meaning, the existing ones) taken instead, following Aristotle , who ‘argues that there is no other way to explain natural generation than by reference to what lies at the end of the process.’
Actually the temporal interaction of two independent processes, random genetic change and the chopping block of natural selection (ie, the fate of species) over immense (or not) time scales is what accounts for evolution.
Given that we can’t beg the question by ascribing the cause to the effect, how do we find a cause, an immediate, proximate cause, for the observation that certain groups prevail and other groups are no longer seen?
A proximal cause for the contrast in the numbers in the corresponding groups is needed.
That cause can only be that in the groups that prevail, in each generation a sufficient number of individuals are born, live to maturity and produce offspring in enough numbers for this process to be replicated in the next generation. Thus we need a self-replicating process, like mathematical induction or perhaps a quine or runaway one, like in computer programming. In the groups that die out, this process is interrupted.
The ultimate causes will be those explored by evolutionary biologists. But rather than a search for an ultimate cause, we want to search for a beneficiary, the agent or actor with objectives whose interests are or are not served by having groups die out, or alternatively by having them continue.
The theory of evolution is saying we don’t need agency to explain whether groups prevail or die out.
I don’t think there is such an agent. There is no purpose behind the self-replication of groups as they persist, proliferate or contract. The process just is. It exists. No why, just a thing.
Groups may benefit from the persistence, proliferation or contraction of their own or other groups, because of their greater desire or greater ability to read the climate of the times. But this search for a beneficiary is not a search for the beneficiaries of persistence, proliferation or contraction. It is a search on a more abstract level for an agent that benefits from the species’ self-replication process itself, and of having life as we know it exist.
But what benefit do groups draw from their persistence? I want to argue that the attention a proliferating species gets at the expense of one that disappears, and the mistaken idea that evolutionary change has goals, together lead to the idea that a proliferating species has the goal of self-preservation, that it has it to a greater degree than the disappearing species and that it is more successful at achieving that goal.
I think that’s a mistake. Self-replication by instances is just something the species does. Making lots of copies of itself leads to self-preservation of the species/population, but self-preservation is arguably not a goal of the species. It is just a consequence of the replication and big population. Inferring goals requires more than a cause and effect relationship, as the theory of evolution, which is not goal-oriented, shows. There needs to be a context of interaction over time.
Deciding AI apps have agency is the same kind of mistake.
The disappearance of family lines does not indicate those lines were failures. It only indicates that there was a last surviving male member. By definition, this member did not have any male children, but a female member might have continued a line of matrilineal descent.
The Shakespeare family line disappeared, after producing William Shakespeare. Does this mean the family line was a failure? Was it its intention that it not disappear?
[ I thought the Shakespeare family line had disappeared, but famous Shakespeares on Wikipedia shows it alive and well. ]
Assuming for the sake of argument that species do not have the goal of persistence by means of self-replication through (a)sexual reproduction of their instances, this creates an anomaly. Species may not have goals, but what about individuals? People have goals. They have ideas about how to reach them and engage in trial and error to reach them. Self-preservation is a goal for most. A more striking anomaly is raised by the sexual activity responsible for preservation of the species? How can it not be the goal of the group to persist if sex is such an important part of life for people? Its biological function is to ensure the self-persistence of the group.
Assuming for the moment that homo sapiens does not have as a goal its self-persistence, let us consider the role of sex in human life and its implications for individual and species goals. What are the goals of people in the area of sex? People have a drive to engage in sex, but specifically for men do they have goals they seek to attain through sexual intercourse? Does a means-end analysis help us understand their sexual activity?
I think the problem biology has defining ‘species’ is connected to this confusion. Are species real or do they exist in name only as imagined concepts like imaginary numbers, or even just the abstract concept of number, perhaps?
Listing possible goals:
Certainly, at each of these points, or stages in sex, the next stage becomes an objective, or a goal. But, I think these events in sex are better considered parts of or stages in sex, rather than the goals of sex, and sex being a means by which the goals are achieved.
For men, orgasms aren’t considered the ends of sex and sex isn’t considered the means by which orgasms are achieved. Men are not considering the purpose of sex when they engage in it. There is no ends-means distinction.
Many men father children. 60% of men in the US are fathers and 75% of men aged 40-49 are fathers. But most men I don’t think engage in sex as a means to an end, for the purpose of having children.
They recognize it as a consequence, foreseen or unforeseen, of sex, but not the reason they engage in it. I think they are surprised rather than relieved when their partners tell them they are pregnant. They may want children, but it is not the reason they are sexually active.
Although the reason some men apparently do have sex is because they want children. Ferdinand_de_Lesseps for example had 12 children in the 15 years between the ages of 65 and 80, before dying 9 years later at the age of 89.
This ‘blind spot’ men have to becoming fathers is no problem for homo sapiens because their biology does the thinking required for them.
The same applies to eating. It’s not really for eaters a means-ends activity, though if you’re hungry, it may be. You know that eating something is a way to stop feeling hungry.
Unless you’re Socrates, who said he ate to live but that others lived to eat.
Read FoodClothingShelter.
What does this inconsistency between the crucial role of eating and sex in realizing the goal of preservation of the species, and the reason men engage in sex and people eat not usually being for this end at all say about the status of the goal?
Certainly, if they have ends in mind for these activities at all, it is for their own benefit or that of the groups to which they belong, not to any abstract desire to ensure homo sapiens, the human race doesn’t die out.
The category mistake of ascribing purpose to evolution and perhaps goals to species is mirrored in the mistake of ascribing goals and agency to AI apps.
Me atBack to AI