The Species-Relativist Argument: The Boundary Line

I’m currently reading Nicholas Agar’s book Humanity’s End. It is an interesting contribution to the ongoing to debate over the merits of human enhancement.

In an earlier post, I introduced Agar’s species-relativist argument against radical human enhancement. The argument proposes that because our use of radical enhancement technologies may lead to the creation of a new species, and because the experiences of that species may not be valuable to us, we are justified in preventing radical enhancement.

I tried to sketch this argument in more detail in the last post. The outline was based on chapter one of Agar’s book. I now want turn my attention to chapter two of the book, which looks in more detail at two of the crucial premises in Agar’s argument.


1. The Questions that need Answering
The premises in question are the following:

• (2) Certain experiences and ways of existing properly valued by members of one species may lack value for the members of another species.

• (3) Activities and policies that promote radical enhancement may lead to the creation of beings that do not belong to the human species.

Both of these premises direct our attention to the boundary line between different species. This boundary line plays a crucial role in the normative and factual components of Agar’s argument.

On the normative side, the claim being advanced is that if a being crosses the boundary line (whether that being is a future version of ourselves or a friend or somesuch) it may then lie beyond the pale of a certain set of values that are significant to us. As Agar puts it, the species concept will “set the boundaries on an important collection of values, and that we should recognize radical enhancement as infringing on these values” (pp. 21-22).

On the factual side, Agar is claiming that the kinds of enhancement endorsed by the likes of Ray Kurzweil, Aubrey de Grey, Nick Bostrom and James Hughes -- these four are the main conversants in Agar’s book -- will create a species boundary between the enhanced and the non-enhanced.

Given the important role that the boundary line between species plays in Agar’s argument, there are four questions that need to be addressed:

(a) What are the available species-concepts and which one does Agar use? 

(b) How is the boundary line defined by Agar’s preferred species concept? 

(c) How might radical enhancement create beings that cross that boundary line? 

(d) Why is that boundary line morally significant?

To be fair to him, Agar is mainly concerned with the first three questions in the course of chapter two. He seems aware of the need to answer the fourth question, but he doesn’t get round to it. Perhaps he will address it in future chapters. Acknowledging this possibility, I still want to pass some comments on this issue at the end of this post.


2. Agar’s preferred Species Concept
There are several species concepts available in the philosophical and biological literatures. John Wilkins -- who has been known to read this blog -- could no doubt regale us for hours around the campfire with tales from history of the species concept. Being somewhat more of a dilettante in this area, I’ll try to be brief and just mention three potential species concepts (from Griffiths and Sterelny, 1999).

First, there is the phenetic species concept. According to this concept, what makes two organisms members of the same species is some measure of their similarity across morphological, genetic or behavioural dimensions. There are problems with this concept. One concerns how we actually measure similarity; another concerns the fact that members of the same species can often be quite different (e.g. due to sexual dimorphism). As a result of these problems, this concept is generally out of favour in biology.

Second, there is the biological species concept (a process-based concept). This concept identifies a species by referring to the processes that create and sustain the species. In particular, to the fact that populations of organisms form reproductively isolated groups. So according to this definition, two organisms are members of the same species if they are part of population that shares and transfers genes.

Finally, there is the phylogenetic species concept (a pattern-based concept). This concept appeals to the tree of life and the relationships between organisms arrayed along the branches of this tree. Accordingly, two organisms are deemed part of the same species if they are part of the same lineage of ancestral and descendant populations and if they share a distinctive trajectory along the tree of life.

One of the important points about the biological and phylogenetic concepts is that they are, broadly speaking, historical in nature: a species is defined by having a particular kind of history. This stands in contrast to the phenetic approach which is based on either clusters of shared properties or, even, essential properties.

Which of the three species concepts does Agar opt for? Unsurprisingly, he goes for the biological species concept. I say “unsurprisingly” since this is the dominant species concept in modern biology. Agar acknowledges that it is not entirely without its own problems, but he can shrug those off. It does not matter to him whether the concept is the biologically or scientifically correct one; it only matters that it performs the normative and factual roles required by his argument. Does it?

3. Reproductive Barriers
Following the biological species concept, reproductive barriers between two populations are the key boundary lines between species. These reproductive barriers can come in a variety of forms. They could for example be physiological -- one organism is physically incapable of mating with another organism -- or psychological -- one organism is repulsed by or lacks the motivation to mate with another organism. They could also be geographical or genetic. Whatever form they take, they must, in accordance with the terms of Agar’s argument, divide one collection of important values from another.

There are problems with these kinds of barriers, particularly when it comes to humans. I want to look at some of these and consider how Agar responds to them.

First, it is not clear that humans are constrained by such barriers. Allen Buchanan (whose book I am also reading) argues, for instance, that technology has effectively allowed human beings to erode almost all reproductive barriers. Even when there are physiological or psychological barriers to reproduction, our technology allows (or will allow) us to combine genetic material according to our whim. Agar is unmoved by such an argument. He writes that:

“Facts about what might or might not happen in a laboratory are actually not directly relevant to decisions about species boundaries. We assign individual organisms to species not on the basis of what happens in human or posthuman laboratories, but according to what occurs in nature.” (p. 22)

This is not an unexpected move, but I worry about it. It’s willingness to disregard what may or may not happen in the lab might contradict something Agar says later on. Also, I think I’m with Buchanan in believing distinctions between the natural and unnatural, especially when applied to humans, are virtually pointless. So much of our lives are bound to our cultural and technological artifices that its difficult to know where nature begins and ends, if anywhere.

A second problem with reproductive barriers is that they may exclude a good number of beings we routinely consider to the members of our species, e.g. the post-menopausal woman, the castrated male, the homosexual, the celibate priest. Many more examples could be adduced. Agar responds to this worry by pointing out that in each of these cases there are past, potential or counterfactual reproductive connections with other human beings: "but for his castration, the male would be able to reproduce"; "the post-menopausal woman could have reproduced in the past"; and so on.

This kind of response is mildly persuasive, but then it raises the question: why can’t we take into account past, potential or counterfactual uses of technology when determining where the species boundary lies? Why not say that but for current technological limitations humans and chimpanzees could form a reproductively connected population? This point is significant. Agar’s belief that radically enhanced humans will in fact constitute a distinct species could arguably be vulnerable to this kind of counterfactual objection.

A third, and for now final, problem with reproductive barriers is that they are not, as Agar notes, all-or-none affairs. He cites the example of the New Zealand bird the black stilt which has been breeding itself to death by mating with the Australian pied stilt. The fact that such interbreeding occurs may suggest that they are not truly distinct species, but Agar argues that the situation is more complex than that. There is still some degree of reproductive isolation between the populations: pied stilts prefer to mate with other pied stilts; and black stilts prefer to mater with other black stilts. It is only when their preferred mating partners are absent that they mate with one another. Given this situation, it might be possible for full reproductive isolation to be reinstalled or for the reproductive barriers to break down completely.

I have no doubt that this is all true, but it seems to create problems for Agar. He argues against radical enhancement by using the precautionary principle. This principle maintains that if there is a risk of particularly bad outcomes arising from a set of activities, we should restrict those activities. But if the creation of reproductive barriers is not an irreversible process, then maybe things are not bad enough to warrant precautionary measures. The force of this criticism depends on how Agar chooses to specify the kind of precautionary principle he is using.

4. Would Posthumans be Non-humans?
The next topic we need to address is Agar’s factual claim that radical enhancement might lead to the creation of a new species. In line with his preferred species concept, this would happen if reproductive barriers were put in place between the normal and the enhanced. He accepts that not all enhancement would create such barriers, he only claims that some might.

What form would those barriers take? Agar notes that there might be physiological barriers created. For instance, genetic enhancements might require the creation of a new chromosome which makes reproduction impossible between those who have the chromosome and those who lack it. Or, if we replace significant parts of our bodies with mechanical or electrical components, we might be prevented from mating with one another.

Agar is not too interested in those kinds of barriers. He is more concerned about the psychological barriers that might be created. He thinks that radically enhanced beings would have little reproductive interest in normal human beings: they would them as something akin to pets. He makes this point through an amusing thought experiment involving a futuristic singles-ad placed by a normal human (“merelyhuman”) that has to compete with that of an enhanced human (“singularityman”). He thinks the psychological boundaries between singularityman and merelyhuman would be more robust and permanent than those between different cultural groups. Indeed, it might be the case that enhanced human beings with indefinitely extended lifespans drop out of the reproductive game altogether.

I won’t challenge any of this here. I have some concerns, expressed above, about the application of the biological species concept to human beings, and I do wonder how much can be inferred from a thought experiment about potential future being, but I’m more interested in the moral significance such reproductive barriers.


5. Why do we care about reproductive barriers?
The species-relativist argument hinges on the claim that by crossing the boundary between species we might be forced to sacrifice or give up something of great value. But if the boundary primarily concerns how we identify and select potential mating partners then I’m having trouble seeing what all the fuss is about.

Now, certainly, I would not deny that reproduction has some value to human beings. But is the value such as to outweigh the potential benefits of enhancement? More precisely, is the value such that a slight risk of losing it is sufficient to warrant taking precautionary measures? One obvious rejoinder to Agar is that even if singularityman is no longer interested in mating with merelyhumans, he might still be interested in mating with fellow singularitarians. And so whatever loss may initially be experienced might be pretty insignificant.

Agar doesn’t address these issues in chapter two. But he’ll need to do it sooner or later. If his argument is to succeed, he will need to convince me that the reproductive barrier is one that we should not cross.