The Red Ape Returns & Willi Hennig Superstar!

20 years ago Jeffrey Schwartz published The Red Ape, making the case that humanity's closest extant relatives within the animal kingdom were orangutans, not chimpanzees. This was contemporaneous with the media hullabaloo around African Eve, so either you could say that Schwartz's timing was perfect, or it was disastrous. Certainly he was swimming against the spirit of the age, and I recall assuming that The Red Ape was a piece of crankery when I first heard about it, its thesis was so outrageous. But historically the idea that humanity descends from Asian apes is less anomalous, and many scientists of Charles Darwin's time did not agree with his model of an African origin for humanity. The skepticism of this idea persisted deep into the 20th century, one reason that Raymond Dart's discovery of Australopithecus africanus was met with less interest than we might assume. Only with further discoveries of numerous early hominids in East Africa by luminaries such as Richard Leaky and Donald Johanson did our probabilistic framework shift so that the unearthing of a wide range of extinct African apes elicit excitement. After all, one of these (or many) are likely to be the possible precursor to the most important species which walks the earth!

But Jeffrey Schwartz is not deterred, and continues to throw stones at the idols of the age. A few years ago he came out with a second issue of The Red Ape. Today I noticed that he has also coauthored a new paper putting forward the model that orangutans are our closest relatives, Evolution of the second orangutan: phylogeny and biogeography of hominid origins:

Our analyses support the following hypotheses: (1) the living largebodied hominoids represent a monophyletic group comprising two sister clades: humans + orangutans, and chimpanzees (including bonobos) + gorillas (collectively, the African apes); and (2) the human–orangutan clade (dental hominoids) includes fossil hominids (Homo, australopiths, Orrorin) and the Miocene-age apes Hispanopithecus, Ouranopithecus, Ankarapithecus, Sivapithecus, Lufengpithecus, Khoratpithecus and Gigantopithecus (also Plio-Pleistocene of eastern Asia). We also demonstrate that the distributions of living and fossil genera are largely vicariant, with nodes of geographical overlap or proximity between Gigantopithecus and Sivapithecus in Central Asia, and between Pongo, Gigantopithecus, Lufengpithecus and Khoratpithecus in East Asia. The main massing is represented by five genera and eight species in East Asia. The dental hominoid track is spatially correlated with the East African Rift System (EARS) and the Tethys Orogenic Collage (TOC).
i-fef40630c3049b385c7113502ac09a0e-500px-Monophyletic.svg.png

Monophyletic simply means that of the set of terminal branches all descend from a common ancestral node. To the left you see the blue highlighting a monophyletic clade: reptiles + birds. Reptiles themselves are not monophyletic, because the class excludes birds, who are descended from the common ancestor of all reptiles. In fact, birds cluster together with crocodiles, alligators, etc., as a monophyletic clade! We don't need molecular biology to know this, it is clear from morphological characters, and has long been so. In fact that's why the hypothesis emerged and became dominant that birds are the last extant lineage of theropod dinosaurs long before the rise of molecular techniques (birds can be shown to be closely related to crocodiles even without the extraction of biological material from fossils).

Though most people do not know what a monophyletic lineage is, the intuition that chimpanzees are our closet relatives in the animal world has sunk deep into the public consciousness. So deeply that Jared Diamond wrote a book which was titled The Third Chimpanzee, implying that humans should be viewed as the third chimpanzee species after the common chimp and the bonobo. The use of these two species as analogs to our own lineage has resulted in warring conceptions of human nature. The authors of The Demonic Males looked to common chimps as an echo of the human ancestor and primal condition, while others suggest bonobos might tell us more. Focusing on orangutans would change the whole playing field and upturn a great deal of reasoning which is predicated on our closer connection to chimpanzees (as well as gorillas).

i-4e4c54d800f84c0754db9c6f63c5954a-figure2a.png

Figure 2a from the paper, to the left, illustrates the main result. I'm not showing you the rest because it relates to Miocene apes and such which I don't know much about. When it comes to fossils Kambiz Kamrani or John Hawks might add some value, but I certainly won't. The way the cladogram was generated is pretty easy to communicate though, they just looked as informative morphological characters (e.g., teeth) and used cladistic methods to generate the most parsimonious tree. The principle of parismony is of course smiled upon in science, and in the context of cladistics what you are trying to explain are synapomorphies, shared derived traits.That is, traits which two lineages exhibit because of common descent, but which other lineages do not because they do not share common descent from the ancestor which originated that trait. Not all traits are created equal when it comes to giving us information which allows one to distinguish between clusters of species which form clades, who we presume emerged over evolutionary time in a contingent manner.

Cladists are almost messianic in their zeal for synapomorphies. There's a reason. A phylogeneticist who collaborated with Allan Wilson's group during the early years of molecular evolution once explained to me that before the cladistic paradigm came to the fore systematics was a total mess with no common currency. Sometimes it was an entertaining mess, such as Ernst Haeckel's famous diagrams, and sometimes an authoritative mess as evidenced by Ernst Mayer's attempts at evolutionary systematics. But arguments always ended with "I said so!" How exactly could one objectively discern right from wrong when intuition ruled the day? Enter Willi Hennig and the cladist revolution, which introduced a genuine system to systematics, so to speak. Using Karl Popper's criterion of falsification and the principle of parsimony a process emerged to clear out the debris of opinion which riddled taxonomy. Modern phylogenetics was born, and the basic ideas were also applied in a molecular context (though from the beginning molecular biologists did not always rely on cladist frameworks, as noted by Schwartz in this paper, nor do they today).

At this point I'll interject something personal. A good friend of mine was trained as a taxonomist, and she told me once that one of the members of her group gave a presentation which did not hue to the strict cladist line. After the talk a cladist approached her privately and berated her for not using "strict Popperian methods." The point here is not the substance but the style. Hennig's ideas added a great deal of rigor and precision, but many people feel that they go too far, and confuse Karl Popper's philosophy of science with science itself. My friend was predominantly a cladist, but her attitude was that there are more things in heaven and earth than only synapomorphies. Taxonomy needed some boundaries in the 20th century, but some cladists are strangling good science in the name of a straightjacket framework.

Jeffrey Schwartz's fixation on morphological characters and dismissal of the sometimes ad hoc and philosophically sloppy molecular methods strikes me as exhibiting this sort of narrowness. The main reason that molecular phylogenetics is viewed as authoritative in evolutionary paleontology is that in the 1970s the fossils and the molecules disagreed, and in the 1980s the fossils came into line with the molecules. There simply aren't that many fossils, and in previous eras the interpretation left much to be desired. With computational methods this is changing, but the reality is that molecular methods in the study of human evolution has withstood the test of time. This simply isn't fleshed out in the paper, which objects to molecular methods in a more general sense.

I don't find Schwartz's objections to the molecular data persuasive. There're the ones I would expect, and I won't go into them because i don't want to get into discussions about the molecular clock and what not. Obviously coalescent theory and such need to be questioned and examined, and models may not be as robust as one would like, but I don't personally find Schwartz's proffered alternative, rejecting most molecular methods as philosophically incoherent, as viable. Morphological characters used in analyses as in the paper above are a small finite subset of the total set of possible characters. In contrast, with the short-term likelihood of a large number of total genomes for both humans and other primates, and computational power to analyzes those genomes, there won't be the problem of methodological bias in terms of characters which are selected for analysis.

I think it is actually of interest to see how human traits and specific genes do not map onto the affinities of total genome content across taxa. In fact, it is interesting to see how this works within human populations. After all, the darkest skinned populations in the world, those of India, Melanesia and Africa, are not a monophyletic group. Rather, Indians and Melanesians cluster with non-Africans generally, light or dark. This is a trivial example, but illustrates that characters which do not map onto general phylogenies are significant and may tell us a great deal of & about the nature of adaptation. My attitude toward the "Red Ape Hypothesis" was that though I dismissed the model I thought the collection of traits where humans resembled orangutans were of interest, note and worthy of further exploration. As it is, Schwartz seems intent on overturning the credibility of molecular systematics itself to stack the odds in favor of his ideas about the origins of humanity. I wouldn't take that bet personally. This might not rank with Fred Hoyle's stubborn adherence to Steady State Theory, but I think the same basic issues are at work. When it comes to models of molecular evolution, mend it, don't end it. We need a diversity of methods, and for many questions fossils offer no answers because of the
small number of remains. The issues relating to this paper are worthwhile because putting a spotlight on the characteristics which can't be predicted by shared descent is going to tell us a lot about evolution, especially adaptation. But these sorts of really radical revisionisms shouldn't lead us to dismissing the traits outside of the main trendline; and yes, genetic sequences are traits too.

Related: For a more detailed review of the paper itself, see The New Scientist.

Note: It seems that to some extent ancient DNA might be a check on some of the molecular evolutionary models which Schwartz contends are not falsifiable.

Cite: Journal of Biogeography (J. Biogeogr.) (2009), Evolution of the second orangutan: phylogeny and biogeography of hominid origins, John R. Grehan and Jeffrey H. Schwartz,

Categories

More like this

The face of Anoiapithecus. From Moya-Sola et al. (2009). One of the most controversial aspects of the whole Darwinius kerfuffle has been the primate's proposed status as "the ancestor of us all." The fossil, named "Ida", has been popularly touted as the "missing link" connecting us to all other…
Good news! The gorilla genome sequence was published in Nature last week, and adds to our body of knowledge about primate evolution. Here's the abstract: Gorillas are humans' closest living relatives after chimpanzees, and are of comparable importance for the study of human origins and evolution.…
In any book about evolutionary anthropology it is almost obligatory to cite Charles Darwin as the person who suspected that our species was most closely related to chimpanzees and gorillas, thus anticipating our modern understanding. In his famous 1871 book The Descent of Man Darwin wrote; In each…
This is the first in an irregular series of basic concepts in science, that I suggested to the Seed Bloggers we might do from time to time. If anyone wants to suggest a revision, because I got it wrong or am unclear, make a comment - this will be revised to make sure it is OK. Clade: this term of…

Good post, Razib. There really seems to be an increasing synthesis between molecular/genetic studies and paleontology (with some developmental stuff thrown in to boot). No one discipline holds all the answers, but what is found in one can be tested with discoveries in another. Another example is the status of whales as derived artiodactyls. That's what the molecular studies said, but paleontologists disagreed until they found evidence that confirmed the hypotheses made from proteins. As I said, it is not so much about fossils being better than genes, but using both to test hypotheses.

science = debate.
The history of science shows many cases where the minority opinion turned out to be correct. We need more data, not more flame.

Dr. Schwartz gave a talk here at WVU last earlier this year. I and another professor asked him questions regarding his conclusions and his opinions on molecular methods. While he ended up talking for nearly 20 minutes, he either never answered the questions posed or gave answers that clearly demonstrated that he didn't have a clue what he was talking about with regard to molecular genetic analyses. It was both frustrating and disappointing. Afterwards, I had dinner with him and even gave him a ride to and from the restaurant. During that time, I continued asking questions. Unfortunately, I believe that he has already decided what he wants the answer to his question to be and will only consider data that supports it. Hence (I suspect), his dismissal of molecular data for ungrounded reasons. This isn't good science. I even attempted to continue the conversation via e-mail but never got a response.

Dr. Schwartz gave a talk here at WVU last earlier this year. I was in the front row. I and another professor asked him questions regarding his conclusions and his opinions on molecular methods. While he ended up talking for nearly 20 minutes, he either never answered the questions posed or gave answers that clearly demonstrated that he didn't have a clue what he was talking about with regard to molecular genetic analyses. It was both frustrating and disappointing. Afterwards, I had dinner with him and even gave him a ride to and from the restaurant. During that time, I continued asking questions. Unfortunately, I believe that he has already decided what he wants the answer to his question to be and will only consider data that supports it. Hence (I suspect), his dismissal of molecular data for ungrounded reasons. This isn't good science. I even attempted to continue the conversation via e-mail but never got a response.

Thank you so much for sharing this story about the genetic similarities between humans and orangutans with your readers. I'd like to invite you to visit the Orangutan Outreach website to learn more about the crisis facing orangutans in the wild.

As I'm sure you are well aware, orangutans are critically endangered because of rapid deforestation and the expansion of palm oil plantations in Borneo and Sumatra. If nothing is done to protect orangutans, they will be extinct in just a few years.

Visit the Orangutan Outreach website to learn how YOU can make a difference!

Thanks for your time...

Richard Zimmerman
Director, Orangutan Outreach
http://redapes.org
Reach out and save the orangutans!
Facebook Cause: http://causes.com/redapes

The history of science shows many cases where the minority opinion turned out to be correct. We need more data, not more flame

yes, but the vast majority of well accepted, tested and utilized theories are less overturned than extended. it is always possible to turn the world upside down, but the general assumption that it is right side up is what you should go with operationally.

How they can be any more confident that our anatomical similarities are primitive retentions than they are about genetic similarities?

It's been about 10 years now, but I took my Physical Anthro course from Dr. Schwartz as a major at Pittsburgh.

I'm intrigued, but not surprised, to see he is still thoroughly dismissive of new techniques.

By Brian Jacobs (not verified) on 19 Jun 2009 #permalink

TGGP,

One reason that morphological characters are not so great to use for phylogenetic analysis is because of the functional complexes phenomenon. Well, first of all, one has to address the availability bias caused by the features usually utilized, since they tend to come from the most common types of fossil remnants found, primary teeth and skull fragments. The genetic code, in opposition, specifies the entire organism. And as stated before, functional groups of anatomical features also tend to evolve together; one such example would be the size of the mandible and anterior dentition in Australopithecus. Another functional group might be brain size, cranial length, breadth, height in hominins. Convergent evolution happens more often than you might think; A. aethiopicus ("Black Skull") and A. boisei both have large cheek teeth associated with heavy chewing of low-quality foodstuffs, and cladistic analysis of that singular trait would place them in a monophyletic group, but using other non-associated traits leads many to believe they are actually paraphyletic cousins. Yet another problem with using morphological features is the need to map qualitative anatomical descriptions to quantitative ones for fine-grained work.

I'm a fish taxonomist; long been a classical morphoologist. But I now think, all being equal, that an analysis of DNA morphology is more likely to approach the One True Tree than a classical morphology analysis. First, it is the DNA which is actually transmitted from generation to generation, so a correct analysis of DNA history is as good as it can get. Secondly, classical morphology is DNA seen through a glass darkly. It is influenced by a number of non-geneological factors, and thus reflects more than relationship.

In the killifish family Rivulidae, there have been two independent DNA trees, and one classical norphology tree generated. The DNA trees are remarkably similar, but there is disagreement with the morphological tree in one area. The DNA trees agree on a lineage which includes fishes which have remarkably diverse morphology. The morphology tree spreads these species around as their morphology suggests. The DNA trees have these species on very long branches within their monophyletic lineage. This suggests to me that diversification in this lineage has proceeded a a very rapid rate in comparison with other lineages in the family. So, if this group is indeed monophyletic, its members have diversified so much that classical morphology does not reveal their true relationship.

By Jim Thomerson (not verified) on 19 Jun 2009 #permalink

TGGP,

One reason that morphological characters are not so great to use for phylogenetic analysis is because of the functional complexes phenomenon. Well, first of all, one has to address the availability bias caused by the features usually utilized, since they tend to come from the most common types of fossil remnants found, primary teeth and skull fragments. The genetic code, in opposition, specifies the entire organism. And as stated before, functional groups of anatomical features also tend to evolve together; one such example would be the size of the mandible and anterior dentition in Australopithecus. Another functional group might be brain size, cranial length, breadth, height in hominins. Convergent evolution happens more often than you might think; A. aethiopicus ("Black Skull") and A. boisei both have large cheek teeth associated with heavy chewing of low-quality foodstuffs, and cladistic analysis of that singular trait would place them in a monophyletic group, but using other non-associated traits leads many to believe they are actually paraphyletic cousins. Yet another problem with using morphological features is the need to map qualitative anatomical descriptions to quantitative ones for fine-grained work.

I've read the Red Ape, and I agree with David, Schwartz does not understand molecular methods. I'm not sure if it's fair to equate him with typical cladists - I think the latter usually understand the methods but impose unreasonable (to the rest of us) philosophical constraints on the analyses. They don't object to all molecular methods, there are plenty of molecular papers in Cladistics after all.

One of the main complaints of the authors seems to be sequence alignment, but let's say you are looking at chimp-human protein sequences, you won't get too many issues with gaps and you are not going to need to do structural alignment.