Marks Review

American Journal of Physical Anthropology 1992

Piltdown: A Scientific Forgery. By Frank Spencer. New York: Oxford University Press. 1990. xxvi + 272 pp. ISBN 0-19-858522-5. $24.95 (cloth).

The Piltdown Papers. By Frank Spencer. New York: Oxford University Press.

1990. xii + 282 pp. ISBN 0-19-858523-3. $65. (Cloth)

[376] Frank Spencer has produced two new works on the Piltdown fraud: one perhaps the most authoritative chronicle to date, the other an edited series of letters from the [377] archives of the British Museum by some of the principals. There are three areas of interest for these works: as documentation of the history of physical anthropology, as documentation of the process of science, and as another contribution to the "whodunnit" genre of Piltdown literature.

The aspect of Spencer's contribution that has received the most notoriety is his suggestion (following Ian Langham) that the perpetrator was none other than Arthur Keith, the leading anatomist/physical anthropologist in England. A brief recap may be in order Most people interested in pinning this tale on someone have followed J.S. Weiner in seeing Charles Dawson as their donkey. Circumstantially, Dawson found almost all the fossils, and after his death no more were found; and he was something of a shady character. On the other hand, Dawson was virtually the only participant who was not a scientist, and therefore is certainly an easy target as an outsider. Through the years, others have cast aspersions on Grafton Elliot Smith, Pierre Teilhard de Chardin, William J. Sollas, Arthur Conan Doyle, and more recently upon some lesser-known suspects: Martin A.C. Hinton, Frank Barlow, and Lewis Abbott.

I confess that I do not find this avenue of inquiry particularly productive, or even particularly interesting. The Piltdown affair should be of considerable interest to us as a community of scientists and as a community of physical anthropologists, but to focus upon it simply as a docudrama is to lose sight of the real significance of the episode. What Piltdown raises, as the archetypal scientific fraud, are questions about the scientific process How does fraud work? What structures exist in science to prevent its detection? Is the critical eye that gives science its vaunted "self-correcting" feature efficient enough? Do the media work in the best interests of the Scientific community when they publicize conclusions that may be poorly supported, and then inflame anti-intellectual sentiment by publicizing its debunking, as it they weren't the main part of the reason it needed to be debunked?

I do not find the case against Keith to be a very convincing one, but neither do I think the case against anyone else is much stronger.. In Keith’s defense we have the fact that he reconstructed the skull poorly and then engaged in a protracted and bitter dispute over the proper manner of reconstructing it. If he in fact perpetrated the fraud himself, it is difficult to imagine him being so absentminded as to have forgotten what it originally looked like! And, certainly, were this simply a ploy (fake it and reconstruct it badly, so that when the fraud is revealed you're in the clear) it would also have the effect of making him look the fool for having reconstructed it so ineptly. Furthermore, it is hard to imagine Keith's motive–not having a vested interest in its authenticity, in that he was not its discoverer or original describer–in publishing several books over his career emphasizing Piltdown's importance, if he knew it to be a fraud. Another recent theory, apparently inspired by Spencer's book, is the inadvertent conspiracy theory of Keith Stewart Thomson (1991), that the original fraudulent fossils were by Dawson; that the later ones were by others who wished to show Dawson they knew what was going on, but without sensationalizing; that it had some elements of a joke (the cricket bat tool); and that it got out of hand with Dawson's death, with no one left to blame except the second batch of tricksters, who would be in no position to let the cat out of the bag.

But this is idle pedantry. What do we learn from Piltdown? Is the "greatest lesson of Piltdown," as per Washburn (1953), "that there never was enough of the fossil to justify the theories built around it" (emphasis in original)? By that criterion, we have to admit that it stacks up pretty well against Taung, WT-17000, and even OH-62, around whom some good yarns, many of them probably valid, have been spun. Do we learn, as per Hooton (1954), first, "that it is unwise to accept current scientific decisions and 'proofs' as final, irrevocable, and conclusive, no matter how authoritative they may sound or look"? (Certainly we knew that independently of Piltdown!) Or, second, "although there may be a few crooks among scientists and their assistants and adherents, the practitioners of science are in the vast majority of cases men who are perfectly honest and so scrupulous in their search for closer and closer approximations to the truth that they will not try to cover up any sin that may have been committed by one of the very few black sheep in their flock, no matter how damaging to the reputation of science that revelation may be"? (Somehow I suspect you'd have a hard time selling that lesson to anyone who has followed the David Baltimore controversy at all carefully.)

For me the most fascinating aspect of the [378] Piltdown case is that William King Gregory returned from a 1913 visit to the BM(NH) noting that "[i]t has been suspected by some that geologically they are not as old at all; that they may represent a deliberate hoax… ‘planted’ in the gravel-bed, to fool the scientists." Extraordinary, whoever suspected such perfidy never came forward, and Gregory’s repetition of this rumor (Spencer attributes the rumor either to Martin A. C. Hinton or James L. Williams) is the only such statement on record. The next closest thing to a suggestion of foul play is Colin Groves being told by Theodore McKown in 1966 that Gerrit Miller told him that some decades earlier he had concluded that Piltdown was bogus, "but had been persuaded by his colleagues not to publish his suspicion on the grounds that without positive proof this would be too serious an allegation of scientific fraud" (Oakley and Groves,


Is it possible that the "self-correcting nature of science is so inefficient that even suspicions about the authenticity of a corpus of work–much less the veracity of conclusions derived therefrom–cannot be safely voiced? Probably. Then, as now, whistle-blowers had something to fear: Gerrit Miller managed to avoid the fate that more recently befell whistle-blower Margot O'Toole. The reason is probably that science is a social, bureaucratic enterprise, and such systems are rarely conducive to whistle-blowers; rather, the bureaucracy protects the miscreants within it, as we saw with painful clarity in the Imanishi-Kari/Baltimore case. Gerrit Miller was right to be scared: People with power and connections use them to protect themselves. It was as true for Cyril Burt as for Trofini Denisovich Lysenko. What we learn from Piltdown is an aspect of how science works; and hopefully we as a community can grow from that new level of self-awareness, and perhaps generate an improved set of norms from it.

Another lesson involves the possibility, raised any number of times, that the association between the skull and the jaw was accidental. Now, of course, given the mosaic nature of evolution, there seems no reason a priori why a derived cranium and primitive mandible could not be associated in the same organism. The argument that they could not have come from the same individual is as applicable to Australopithecus afarensis as to Eoanthropus dawsoni. Indeed, I would suspect that such a false argument is related to the orthogenetic views of human evolution popular at the time–in which progress "to ward" humanization involved the entire body evolving together. In such a light, of course, the "associationist" view of Piltdown might be the more enlightened, and thereby

deserved to win the day.

However, even granting that a reasonable argument against the association of the Cranium and teeth could be made, the apparent association would have had to be accidental, not calculated. Miller wondered in print that "deliberate, malice could hardly have been more successful than the hazards of deposition in so breaking the fossils as to give free scope to individual judgement in fitting the

parts together." But the breakage pattern of the fragments never an issue: it was that an ape's jaw was there together apparently with a human's cranium, and, if you didn't believe it was real, it had to be an accident, for there was no other alternative. David Hull (1988) points out that even though fraud and error lead to the same conclusion (wrongness; blind intellectual alleys for the rest of the

community), scientists are more willing to tolerate error than fraud. In other word:,. it is just better in science to be stupid than dis honest. That is why the initial defense in all scientific fraud accusations is that it was accidental, not deceitful, for we all make errors and are consequently likely to be sympathetic. Again, Piltdown represents science practiced-as-normal, for whatever problems

arose had to be honest problems, until the smoking gun of 1953, when almost everyone involved was dead anyway.

One reason that Piltdown held as much sway as it did was that its supporters in voked the principle of replication. Was the ape's teeth and human's cranium a fluke or a real association? If those again were the only two options, then the discovery of Piltdown 2 settled it (Washburn, 1979). But obviously, there is a broad spectrum of reasons why scientists might get the same results twice, only one of which is that the results are, "right." Obviously, if fraud was committed

once, it could be committed twice. Alternatively, scientists often see what they are looking for, as in the case of the 47th and 48th human chromosomes (Kottler, 1974). Or the same experimental artifact can occur independently, as in polymerized water (Franks, 1980). Or scientists, who often have interests other than simply discovering truth, may bias their results to support ide ologies, pet theories, friends, colleagues, or superiors. Again, in the context of how sci[379]ence works, Piltdown provides an illustration of normalcy, not deviance. The deviance is that it happened to have been chicanery but the activity of the scientists involved was normal, if not paradigmatic. A notable irony is that the fluorine test, which finally was taken as proof of fraud, once fraud was suspected, was actually a second try and gave the expected results: The first time it also gave the expected results, that the jaw and skull actually belonged together chronologically (Oakley and Hoskins, 1950).

Perhaps the most bizarre aspect is what Piltdown reveals about the judgments of experts. I do not mean this in the sense that Hooton raised the question, as referring to the interpretations of the experts, which are generally sound but always subject to revision. Rather, I mean it in a more primary and fundamental sense. The dentition of Piltdown is now recognized as that of an orangutan, and orangs are dentally in some ways more similar to humans than to chimps and gorillas. And the diagnostic occlusal surface of the molars had been filed away. Yet somehow, the ape invoked as the source of the jaw by nearly all the "non-associationists" was a chimpanzee! Miller actually named this new chimp, known only from its jaw, as Pan Vetus. And, for all the dental expertise on the part of the early critics of "association" (Waterston, Boule, Matthew, MacCurdy, Hrdlicka), only Gregory (1914) saw dental similarities to an orang (in the canine). Curiously, Gregory also maintained that the molars are chimp-like (1916) but shortly before the discovery of fraud he wrote "In particular, [the Piltdown molars] rather closely resemble the worn crowns of certain orang molars" (1951:497). Wouldn't it be nice to know on what basis he revised his opinion? Spencer attributes the first recognition of orang-ness in the molars to a German anatomist named Heinz Friedrichs in 1932, and more, widely publicized by his professor Frank Weidenreich (1936).

In addition to the experts who apparently could not tell orang-utan teeth and jaw fragments from their chimpanzee homologs, we have Hrdlicka's certainly informed judgment: "It resembles more or less in a number of points the jaws of the chimpanzee, but it differs from these in a whole series of points of importance, such as the form of the notch, type of coronoid process, subdued musculature. markedly reduced internal massiveness of body especially near symphysis; and in the most important characteristics of the teeth, namely, height of crown, height of enamel, nature of 'cingulum' and stoutness of cusps–in all of which features it is nearer or like human" (1922:346). Spencer notes that Hrdlicka (unlike Gerrit Miller) examined the original specimens on a visit to England in 1922. How did he find human characteristics in an orang-utan's mandible? (And that would not be the last time the difference in enamel thickness between humans and orangs on the one hand, and chimpanzees on the other, would provide a rich source of confusion to paleoanthropologists!)

Finally, Piltdown raises a general question about the literature. Once a piece of work has become problematic, how do we exorcise it from the literature? On the one hand, we want to make sure that we are comprehensive in our discussion of the data and cast our net widely; thus Piltdown was still part of the story long after Weidenreich (1936) was noting that the fossil record itself belied Piltdown's significance. On the other hand, one has to make decisions about what to include, and both Hrdlicka (1930) and Hooton (1931) chose not to include Taung in their reviews of paleoanthropology.

Where were the Literature Police? Piltdown, being the paradigm fraud, has never directly confronted the problem Conway Zirkle articulated in 1954: that Paul Kammerer's work on Lamarckian inheritance in midwife toads had been rediscovered by a generation of workers who did not know (and by some who did not care) that the work was fraudulent. But how do we monitor the literature to insure not only that fraudulent results are kept to a minimum, but that once a piece has been debunked, it ceases as well to be cited as if it had been competent and honest?

What Spencer's books yield is the best opportunity for us to study the Piltdown case and learn about how we, as scientists, act and think. Perhaps when we as social scientists study ourselves more carefully as a social community, we will come to appreciate the real significance of the affair. It does not show how unique physical anthropology is as a science–it shows quite the opposite, how much like other sciences it is. But, it gives us, as a science a little more introspective than others, an opportunity to reflect on it and to grow from it.

A final thought: If the fraud had been discovered promptly, and the perpetrator identified, what would have happened? Paul Kammerer blew his brains out, but that [380] seems to be an exceedingly rare fate for data falsifiers. Instead, they tend to claim 1) it was all a mistake; 2) they're being persecuted; 3) it is simply a difference of interpretation; or 4) they got the "right" answer, so ultimately there isn't really a problem. The success of these claims again depends on the ease with which the bureaucratic nature of science can be exploited. Frankly, I would bet that if the perpetrator of Piltdown had been nailed, s/he would have gotten off Scotfree, especially if it had been someone prominent such as Arthur Keith (who was already a Scot)....

Jonathan Marks

Departments of Anthropology and Biology Yale University

New Haven, Connecticut


Literature Cited


Franks F (1980) Poilywater. Cambridge, MA: MIT Press.

Gregory WK (1914) The dawn man of Piltdown, England. Am. Mus. J. 14:188-200

Gregory WK (1916) Note on the molar teeth of the Piltdown mandible. Am. Arithropol. 1 8:3s4-387.

Gregory WK (1951) Evolution Emerging. New York: Macmillan.

Hooten EA (1931) Up From the Ape. New York: Macmillan.

Hooten EA (1954) Comments of the Piltdown affair. Am. Anthropol. 56:287-289.

Hrdlicka A (1922) The Piltdown jaw. Am. J. Phys. Anthropol. 5:337-347.

Hrdlicka A (1930) The Skeletal Remains of Early Man: Washington DC: Smithsonian

Miscellaneous Collections, 83.

Hull DL (1988) A mechanism and its metaphysics: An evolutionary account of the social and

conceptual development of science. Biol. Philos. 3:123-240.

Kottler MJ (1974) From 48 to 46: Cytological technique, preconception, and the counting of

human chromosomes. Bull. Hist. Med. 48:465-502.

Oakley KP, and Groves CG (1970) Piltdown Man: The realization of fraudulence. Science


Oakley KP, and Hoskins CR (1950) New evidence on the antiquity of Piltdown man, Nature


Thomson KS (1991) Piltdown man: The great English mystery story. Am. Sci. 79:194-201.

Washburn SL (1953) The Piltdown hoax. Am. AnthropoL 55:759-762.

Washburn SL (1979) The Piltdown hoax: Piltdown 2. Sicence 203:955-958.

Weidenreich F (1936) The mandibles of Sinanthropus pekinensis: A comparative study.

Palaeontol. Sin. Series D7.

Zirkle C (1954) Citation of fraudulent data. Science 120:189-190

Back to the

Back to the

Back to