What It Really Means To Be 99% Chimpanzee
Jonathan Marks
Department of Anthropology
University of California, Berkeley
Presented at the Annual Meeting of the American Anthropological Association
Biological Anthropology Today:
Topics For Non-Biological Anthropologists
(Presidential Symposium)
Saturday, November 20, 1999

I want to do something that is classically anthropology. I want to take a well-known natural fact and show it to be a construction of the social and cultural order, and in that capacity, in need of de-construction. I’ve worked on this question both as a geneticist and as an anthropologist. That is to say, both as a technologist, using machines with flashing multicolored lights; and as an ethnographer, extracting meaning from words and acts.

So let’s begin with the natural fact: Humans are genetically very similar to chimpanzees and gorillas. If you look at the structure of their chromosomes, for example, you find a few minor rearrangements and only one translocation differentiating them [Slide: 5/17]. This is some work I did a few years ago, in which the DNA from human chromosome 5 is colored pink, and the DNA from human chromosome 17 is colored green; here on the gorilla’s chromosomes, rather than 2 pairs of pink and green chromosomes, you see two pairs of chromosomes each partly-pink and partly-green. The gorilla’s DNA is very similar to the human structurally and functionally, but slightly reorganized geographically.

In fact, however, the overwhelming genetic similarity between human and apes was known in the 1920s.  It was invoked in the context of the Scopes trial in H. L. Mencken’s literary magazine, the American Mercury – blood serum reactions standing for genetics, and humans and apes being more similar by that measure than horse and donkey. That article, of course, failed to draw the conclusion that we "are" apes, in spite of those data. That inference would have to wait for the rise of molecular reductionism in the 1960s.

And the classic articulation of this idea was by a chemist named Emile Zuckerkandl, at a Wenner-Gren conference called Classification and Human Evolution in 1962. Zuckerkandl was so struck by the similarity of the protein sequence of hemoglobin in human and gorilla – only one difference in 287 amino acids – that he could proclaim, "from the point of view of hemoglobin structure, it appears that gorilla is just an abnormal human, or man an abnormal gorilla, and the two species form actually one continuous population."

Well, of course, as the distinguished paleobiologist George Gaylord Simpson was obliged to point out in quick response, a gorilla is not an abnormal man; it’s a gorilla. Gorillas and humans do not form a single population any more than gorillas and kangaroos do.

Humans and gorillas are diagnosably distinct anatomically, behaviorally, mentally, ecologically, demographically – effectively any way that you can compare them. And if you can’t tell the human from the gorilla because you’re looking at their hemoglobin, just look at something else!

In spite of the commonsensicalness of that observation, thirty-odd years later Zuckerkandl’s view is far more widespread than Simpson’s. Hemoglobin has proven to be a synechdoche for genetic similarities, whether measured by narrow but precise DNA sequence or by broad and crude DNA hybridization.

Humans, chimpanzees, and gorillas are within two percentage points of one another genetically, even if not obviously any other way. And this revelation, which has been known since the 1960s, and acknowledged in one form or another for decades previously, now forms the explicit basis of a spate of popular science books telling us that, for example: 

Apes merit human rights, perhaps 98%, in spite of the fact that they aren’t human, we don’t allocate rights on the basis of genetic distance, and we can’t even guarantee human rights to humans.


We are less than a 2% variant chimpanzee, which is simply Zuckerkandl’s argument, although without acknowledgment here.


My personal favorite: We are over 98% demonic males, because chimpanzees are, if chimpanzees are.

The point is that there is little of scientific value here, but lots of cultural value. Not least of which is the cover of the latter book, by a Harvard biological anthropologist, illustrating the point with a black guy and a gorilla, and being, unfortunately, a broadly more influential representative of this field today than the collectivity of this morning’s distinguished panel. Remember that.

Nevertheless, as Zuckerkandl took it in 1962, regardless of whatever differences appear to exist, the genetic comparison is all that matters, and it overrides everything else; it transcends the overall biological relationships.

In other words, the 98% similarity is not so much a scientific discovery as it is a cultural discourse of heredity. One, of course, that we are familiar with in the era of the purple prose of the Human Genome Project, aka the holy grail and the book of life.

The first issue I wish to address about the genetic similarity is its frequent representation as a sort of paradox.  Given a human and chimpanzee, you can easily tell them apart, but given only their DNA, you can’t tell them apart.

But there is a bias of history here. We’ve been studying chimpanzees for 300 years, but DNA sequences for barely 20 years. We are far more familiar with apes than we are with DNA. Consequently, the appropriate way to compare these data is not to contrast the genetic and anatomical comparisons through modern eyes, but to compare the genetics today with the anatomical comparisons when those were as new and as exciting as DNA comparisons are today.

And what you find is that the leading scholars of the 1700s, the leading philosophes, were struck by the overwhelming physical similarity of ape and human. Rousseau and Monboddo were struck by the humanness of the ape, and declared it to be a variant human.  Linnaeus famously classified the apes as bothHomo troglodytes or nocturnus – a different kind of human – and as Simia satyrus, a different kind of monkey.

So the point I wish to make is that the paradox of the anatomical difference and the genetical similarity is illusory – it’s an artifact of the intellectual history of comparing. How familiar we are at the turn of the millennium with the physical differences and how unfamiliar we are with the whole notion of genetic difference.

Let me illustrate that a different way.

One of the central questions in contemporary biology is the manner in which the one-dimensional DNA sequence is operationalized to construct a four-dimensional being, a developing, an evolving – in the original sense of that term – organism. That very incommensurate dimensionality renders the comparisons themselves incomparable.

To assess the percentage of difference between two linear DNA sequences is easy. Two assess the percentage of difference between two developing bodies is meaningless. And it’s meaningless for a specific reason: it lacks context. What molecular genetics brings is the ability to compare two organisms in an entirely decontextualized framework.

Some of you older folks may remember Allan Wilson’s research group wrestling with this issue in the 1970s: Given that human and chimp were genetically so similar, how physically similar were they? And they treated it as a statistical problem, and largely unresolved – would a human and chimp look so different to a frog, they asked.

If you want to know how similar a human and chimp DNA sequence is, you simply add up the matches. But if you want to know how similar a human and chimp are, the obvious question is: Compared to what?

Compared to a starfish, let’s say , humans and chimpanzees match in having bilateral symmetry, a central nervous system, a skeleton – bone for bone, muscle for muscle, organ for organ, humans and chimps match right down the line. They’re not 98% physically identical, they are 100% identical, if your frame of reference is an echinoderm.

The body and DNA paradox thus isn’t a paradox at all. The pairwise DNA comparison is an empirical question; the pairwise body comparison is nonsense. In fact, it could be the silliest thing I ever hoid.

The second point I’d like to make is complementary to that. It’s about similarity.

We are accustomed to imagining scales of similarity ranging from 100% similar – that is to say, identical – to 0% similar, that is to say, totally different. This is the conceptual framework within which we interpret the 98.6% or whatever similarity of human and ape. They are really, really similar, almost identical.

But in fact, DNA similarity is not structured in quite that way. There are, as everyone knows, only 4 bases in DNA. And this places an odd statistical constraint on the comparison of sequences. No DNA similarity at all – that is to say, two random sequences that share no common ancestry – are still going to match at one out of four sites. In other words, the zero mark of a DNA comparison is not zero percent similar, but 25% similar.

Once again, the DNA comparison requires context to be meaningful. Granted that a human and ape are over 98% genetically identical, a human and any earthly DNA-based life form must be at least 25% identical. A human and a daffodil share common ancestry and their DNA is thus obliged to match more than 25% of the time. For the sake of argument let’s say 33%.

The point is that to say we are one-third daffodils because our DNA matches that of a daffodil 33% of the time, is not profound, it’s ridiculous. There is hardly any biological comparison you can make which will find us to be one-third daffodil, except perhaps the DNA.

In other words, just as Simpson argued in the 1960s, the genetic comparison is exceptional, not at all transcendent. DNA comparisons overestimate biological similarity at the low end and underestimate it at the high end – in context, humans are biologically less than 25% daffodils and more than 98% chimpanzees.

The focus on base-pair mismatch itself is misleading, for it encodes a number of archaic assumptions about genetics and evolution. In fact, it ignores what is quite possibly the most significant development in biology in the last quarter-century – namely, the complexity of genome structure.

If humans and chimpanzees are over 98% identical base-for-base, how do you make sense of the fact that chimpanzees have 10% more DNA than humans? That they have more alpha-hemoglobin genes and more Rh bloodgroup genes, and fewer Alu repeats, in their genome than humans?  Or that the tips of their chromosomes contain DNA not present at the tips of human chromosomes?

Obviously there is a lot more to genomic evolution than just nucleotide substitution. But the percentage comparison renders that fact invisible, and thus obscures some of the most interesting evolutionary genetic questions.

Once you recognize that there are easily identifiable differences genetically between humans and chimpanzees – the presence of terminal heterochromatin is 100% diagnostic – you can begin to see that the pattern of relationships between the species is actually the same genetically as anatomically. Humans and chimps are simply very similar to, yet diagnosably different from, one another.

The last point concerns the one way in which one could say with some legitimacy that "we are apes" – as, for example, gene enthusiast Richard Dawkins does. That is, phylogenetically. We fall within a clade that incorporates chimpanzees, gorillas and orang-utans. That is a great ape clade, it subsumes us as well, and therefore, we are great apes too.

Now this may sound persuasive, but once again, it says more about culture than about nature.

The processes of evolution sometimes produce highly divergent relatives. A classic example is the evolutionary divergence of the birds from a group of reptiles.Birds are more closely related to crocodiles than crocodiles are to turtles, in spite of the latter two being green scaly crawling disgusting creatures. We commonly recognize the species "left behind", as it were, being similar to one another, although sharing ancient features. They are simply the group constituted by not having evolved what the birds did.

The goal of cladistics is to purge such groups from the formal literature. Nevertheless in the informal literature, reptiles are those green scaly things that didn’t evolve into birds – in other words, they are defined by the divergence of birds from them.

Another well-known paraphyletic category is fish.  Coelacanths and tunas are both scaly, gilled, swimming things, but the group that gave rise to the tetrapods was closely related to the coelacanth, based on things like the structure of its fins. This means that coelacanths and tetrapods are more closely related to one another than either is to a tuna.

But more significantly, since humans are tetrapods, it means that humans fall into that paraphyletic category we call fish.

In other words, we are phylogenetically apes, but only in precisely the same sense that we are phylogenetically fish.

Doesn’t sound quite so profound now, does it?

The claim that we are apes is not a fact of nature, but an artifact of the way we organize and divide nature.

As a consequence, that claim has a lot more to do with the work of Durkheim and Mauss than with Darwin and Huxley. This is an obvious site for a biological anthropology – a field both anthropologically-informed and biologically meaningful. And it’s vacant at the moment.

As many of you know, this subfield has virtually ceased to exist as a curriculum at some major departments. The Biological Anthropology Section is one of the smallest sections in the AAA. The textbooks are almost devoid of anthropological content, and uniformly give more emphasis to seed-color in peas than to the denial of health insurance on the basis of a genotypic diagnosis in people.

Bio-anthro today is the most racialized, the most racist, the most reductivist and the most essentialized of any organized, ostensibly anthropological, endeavor today.

Fortunately there is some reason to be optimistic. This morning’s panel, for one thing. For another, a recent Wenner-Gren conference explored the question of the relationship between anthropology and genetics, and took a crucial step toward interrogating a part of biological anthropology, anthropological genetics, and situating it as an anthropological discourse. . And for a third, let me call your attention to a paper on Iceland in the newest issue of Anthropology Today by Gisli Palsson and Paul Rabinow, who call specifically for "a molecular anthropology that includes scientific, technological, political, cultural, and ethical dimensions."

Our place in nature is under-determined by genetic data. Our place in nature is a contested site on the boundary of humanness and animalness, beast and angel. To make sense of the data requires an anthropological eye and biological knowledge, for its meaning is technologically constructed and ideologically situated.

The anthropological content of biological anthropology is as endangered as the mountain gorilla. It remains to be seen whether it can be rescued from extinction.

Jonathan Marks
Department of Sociology and Anthropology
University of North Carolina at Charlotte
h o m e
email: jmarks@email.uncc.edu
phone: (704) 687-2519
fax: (704) 687-3091