Header image

john hawks weblog

paleoanthropology, genetics and evolution

Photo Credit: Pre-Clovis Gault Assemblage artifacts. Thomas Williams et al. (2018) CC-BY-NC

Lida Ajer, early modern human remains in island Southeast Asia

In August of last year, Kira Westaway and colleagues published new dating results from Lida Ajer, Sumatra: “An early modern human presence in Sumatra 73,000–63,000 years ago”. The cave has a fossil-bearing breccia that was excavated by Eugene Dubois during the late 1880s.

Dubois lost interest in excavating Sumatran caves because he believed the fossils were too recent to document the “missing link” he sought. In 1948, D. A. Hooijer described some of the orangutan teeth from Dubois’ collection from Sumatra, along with two human teeth from Lida Ajer. He concluded that these two teeth, an upper central incisor and a likely second upper molar, were indistinguishable from modern human samples.

Westaway and colleagues were able to relocate the Lida Ajer cave by using Dubois’ notes, while they have been unable to find other caves that Dubois investigated.

Gilbert Price, one of the team members, has put a short video on YouTube giving some context of the team’s initial reinvestigation of the site.

It is great to be able to see the site’s surroundings and a bit of the fossil context in this way.

Hooijer had speculated that the fossil assemblage might have originated as a porcupine accumulation. The collection is nearly entirely teeth and some of them–including the two human teeth–bear evidence of porcupine gnawing. Westaway and coworkers agree with this assessment and further suggest that the fossil-bearing breccia may have flowed into its current location as a mass and later lithified.

The conclusion that the teeth are modern human rather than some archaic human form or Homo erectus is clear enough.

The Lida Ajer teeth are smaller than fossil orangutans and east and southeast Asian Homo erectus/archaic Homo sapiens (Extended Data Fig. 3). They show greater affinity to east Asian Late Pleistocene H. sapiens than to southeast Asian Late Pleistocene to mid Holocene H. sapiens (Supplementary Tables 3, 4). The relative enamel thickness of the incisor is most similar to mean values for modern humans, and exceeds the extant orangutan range (Extended Data Fig. 4 and Supplementary Table 5). Relative enamel thickness of the molar is intermediate between mean values of living humans and living and fossil orangutans. Discriminant function analysis of molar enamel–dentine junction morphology classifies it as H. sapiens (Extended Data Figs 4, 5, 6). Both teeth have a simple external morphology typical of H. sapiens. They lack traits that characterize east and southeast Asian H. erectus/archaic H. sapiens and Homo floresiensis. Furthermore, derived H. sapiens features are found in both teeth, such as incisor double shovelling (Supplementary Information). The combination of their small size and external and internal morphology demonstrates that they are anatomically modern Homo sapiens.

Westaway and coworkers have added substantially to Hooijer’s description by adding the microCT observations, including the enamel-dentin junction (EDJ) and data on enamel thickness. These teeth do not differ from the teeth of Late Pleistocene East Asian human samples, which include the very early teeth from Tam Pa Ling, Laos, and Liujiang, China, along with somewhat later Chinese, Vietnamese, and other archaeological sites.

Based on the authors’ model of site formation, they assume that the breccia deposit formed in a short period of time and does not sample fossils from across substantial geological time. They obtain estimates of geological age from the breccia in four ways: Uranium series dating of flowstones and other speleothem samples, uranium series dating of fossil teeth, ESR dating of the fossil teeth, and red thermoluminescence dating of quartz grains in the breccia.

All of these methods arrive at a similar picture. The U-series dates on the fossil teeth provide minimum ages, and these are all over 40,000 years ago, and most age estimates including ESR and red TL within the range from 80,000 to 60,000 years ago. The flowstones that bracket the top and bottom of the authors’ stratigraphic profile of the breccia are 71,000 +/- 7 and 203,000 +/- 17,000 years old. None of the U-series dates from within the breccia are nearly as old as the underlying flowstone.

As I read the paper and supplementary information, I kept looking for documentation to convince me that Dubois actually collected the two human teeth from within the breccia in the stratigraphic position shown by the authors. To me, the provenience of these two teeth is a critical question. I don’t have any specific reason to doubt that they come from within the breccia where they are reported, but with any site excavated so long ago, we do need to ask the question. The human teeth have no direct date, look like those of modern humans, while lacking the mineral staining seen on some of the faunal remains that they have pictured.

The cave has no evidence of stone tools or other artifacts. The authors discuss this in their supplementary text:

No evidence of archaeology was observed in the cave or surrounding region. Hooijer did “not consider the two teeth described above as evidence of human inhabitation in the prehistoric Sumatran caves”. Hooijer’s comments were made in reference to what he considered a non-occupation cave. No excavations have been conducted in the front of the cave, and whether or not prehistoric peoples used the cave remains to be demonstrated. Nevertheless, the assemblage reflects the fauna that were present on the landscape, humans included. Furthermore, we note the presence of human remains but an absence of archaeological evidence is a common feature in other Asian caves, such as Thum Wikin Nakin in Thailand, Punung in Java, and Fuyan in southern China.

It’s typical to hear that archaeological sites vastly outnumber sites with hominin remains. From that generalization, it might seem reasonable to suppose that archaeological sites document the dispersal of human populations with more fidelity than hominin bones and teeth.

But most time intervals are poorly represented by archaeological sites in most regions of the world. Southeast Asia and island Sundaland have a very sparse archaeological record, even though hominins were likely there in large numbers through most of the Pleistocene.

Archaeologists and anthropologists have expressed many ideas about that sparse record. Some of the area was forested during large spans of time, and many anthropologists have expressed the idea that hominins could not penetrate forested habitat effectively until the last 20,000 years.

Lida Ajer is the earliest hominin discovery in rainforest anywhere in the world [but see UPDATE below]. I don’t think that’s because hominins weren’t in these forests, I think it’s because these areas have not been explored with the intensity of lake edge sites and caves in arid and temperate habitats.

During the last fifteen years, there have been remarkable discoveries in island and mainland Southeast Asia, and I expect that to continue as people keep going.

UPDATE (2018-01-08): I have an e-mail from John de Vos informing me that I should investigate the Punung site on the island of Java.

Punung has a rainforest fauna dated to around the last interglacial (in other words, around 120,000 years ago) and a single human premolar, which is similar to modern humans in its measurements.

I’ll follow up with more information on Punung in a separate post.

The surprising connectedness of human genealogies over centuries

A new article by Adam Rutherford in Nautilus may be a good one for students in my genetics course this upcoming semester: “You’re Descended from Royalty and So Is Everybody Else”. The article is an excerpt of Rutherford’s book, A Brief History of Everyone Who Ever Lived: The Human Story Retold Through Our Genes.

Chang’s calculations get even weirder if you go back a few more centuries. A thousand years in the past, the numbers say something very clear, and a bit disorienting. One-fifth of people alive a millennium ago in Europe are the ancestors of no one alive today. Their lines of descent petered out at some point, when they or one of their progeny did not leave any of their own. Conversely, the remaining 80 percent are the ancestor of everyone living today. All lines of ancestry coalesce on every individual in the 10th century.

When you get up to around a trillion potential lines of genealogy, you have room for some surprisingly long-distance connections. Still, what people tend to ignore about this kind of logic is that incredibly tiny fractions of the genealogical tree are not different from zero when it comes to DNA ancestry.

The argument is based purely on number logic, and is not falsifiable by any empirical observations, short of impossibly complete genealogical knowledge.

In my opinion, we should be a bit more conservative (as Rutherford also reflects in this excerpt). But still, it’s very likely that common ancestors of all living humans have lived within the last 2000 years.

By the same logic, of course, every living human is a descendant of Neandertals. And every Neandertal that was an ancestor of the last Neandertal-modern hybrid was also an ancestor of all of us.

MOOCs after five years

Five years ago, I was just starting to prepare a massive open online course (MOOC). That course development would be an 18-month adventure for me.

Our team worked with the concept that technology can bring students who are learning outside the classroom even closer to the course content than students within the classroom. For a course in human evolution, that meant traveling to the field, bringing students to the sites where fossil hominins have been found. And it meant allowing real scientific experts to speak for themselves.

The course ran in the spring of 2014, with the title “Human Evolution: Past and Future”. People today can watch many of the course components on my YouTube channel, including some remarkable interviews and site visits.

I learned a lot while teaching the MOOC, which has helped me in many ways to develop other forms of public engagement. Many of the video presentations from my MOOC have been even more successful on YouTube than they were in the course with its 40,000 registered students.

Diane Lorillard and Eileen Kennedy, specialists in digital technologies and education, have a post in the Times Higher Education on massive open online courses (MOOCs) from today’s perspective: “Moocs can still bring higher education to those who really need it”.

Their post recognizes some of the ways that the MOOC scene has changed since 2012, and ends with a framework worth sharing:

Viable solutions to making online learning affordable and sustainable require an understanding of the true costs. What are the costs of a video to be shown over 10 runs of a course? What are the costs of facilitating a discussion for each of those 10 runs? How large can an online tutor group be? What combination of activities will produce the best experience for the learner?
More research on this activity-based costing approach will enable us to plan costs realistically in relation to the quality of the learning experience provided.
Online learning at scale has the potential to transform access to quality higher education. It also has the potential to transform what it means to teach in higher education. The question now is how can we make sure that this transformation is productive and sustainable for the future of higher education for all.

Initially, as universities began developing these courses in 2012, they were not driven by real sustainable motives. In the first couple of years, MOOCs were a way for premier universities to compete for status in a new area that was getting a lot of press.

What may have been less visible is that MOOCs were also a way for specialists in educational technology to test new platforms and methods of delivering teaching and interaction online.

My MOOC, for example, enabled the University of Wisconsin to train a large group of educational technology specialists in new technologies, at the same time that it gave more than 40,000 people the opportunity to learn about human evolution from the field.

Many people who were great fans of my MOOC ask, why don’t we do it again? I’ve been asked by other universities, also, to help develop MOOCs on similar subjects.

The truth is, doing a MOOC in anthropology well requires many people to invest time in those personal interactions, on the virtual forum, message boards, and giving personalized feedback on assignments. That kind of interaction may be less necessary in very technical fields, as very popular MOOCs in artificial intelligence and programming have demonstrated.

But to me, a course in anthropology is about the human interaction. That human interaction is labor-intensive. It’s hard to do at a large scale without sustainable funding.

Many institutions still see MOOCs as an inexpensive way to do education at a large scale. That’s not realistic in anthropology. A great MOOC may be relatively inexpensive for the scale, but it is not without substantial ongoing cost.

Still, MOOCs have a serious benefit: A huge population of people in Africa, South Asia, Southeast Asia, and rural areas of many other countries are underserved by local and regional educational institutions. MOOC-like courses can reach people where they live, on the devices that they use.

I dream of bringing those populations into the study of human evolution, where new discoveries are being made. Tomorrow’s generation of paleoanthropologists must represent the areas where tomorrow’s fossil discoveries will be made.

How can we empower people to be a part of this science? To me, that’s the big problem. My instinct is that we can build communities to make this kind of learning possible for people around the world.

Bonobo preferences: dominance over cooperation

An article in Current Biology by Christopher Krupenye and Brian Hare suggests that bonobos may have a social preference for individuals who wear their dominance on their sleeve: “Bonobos Prefer Individuals that Hinder Others over Those that Help”.

This has gotten a good amount of press attention this week, contrasting the bonobos with humans in terms of cooperativeness and prosociality. It’s an instance where the bonobos seem to be acting more like chimpanzees than the usual highly prosocial bonobo stereotype.

Michelle Rodrigues is a primatologist at the University of Illinois who has read the new study in detail and wrote a very helpful summary: “Human infants prefer helpers, but adult bonobos prefer hinderers”.

From the post:

So what do these results tell us? I think Krupenye and Hare have nicely demonstrated that adult bonobos in a sanctuary setting prefer "hinderers" or dominant individuals. However, these results don't hold for the younger bonobos (ages 4-9). And that's where the comparison to humans fall short. We can't compare human infants to adult bonobos, and then conclude that this is a species difference. I suspect that this may be an age difference in both species, though there also may be greater variation depending on culture, personality, socialization, etc.

It is a fascinating question to what extent apes may be enculturated by their exposure to a dominance hierarchy and the behavior of older individuals in their social groups. We know that humans are highly plastic in their development of social preferences. But we don’t know whether the complex landscape of social interactions may have “attractors” that may affect or reinforce cooperation versus competitiveness and dominance hierarchies.

Link: Expedition journal from Niah Caves

I was really pleased to see a post by Darren Curnoe recounting his team’s recent field season in Niah Caves in Borneo: “We Found Evidence of Early Humans in the Jungles of Borneo”.

Over a period of three weeks, we dug through what we believe to be around 20,000 years of human history. We uncovered several human bones, the remains of large mammals (probably deer and wild cattle) and marine oyster shells indicating a period of seafood meals. Stone tools and charred rocks were also unearthed.
It was exciting and a little bit daunting to be digging at Niah Caves, given its place in both the history of archaeology and more broadly of humankind.

Really great to see more teams keeping expedition journals, which are such powerful ways of communicating the real-life decisions and processes of archaeological science. Curnoe’s “Daily Dig Diary” on YouTube (indexed on his Facebook page) was an informative way to follow archaeological research as it happened.

Link: Implications of an 11,500-year-old genome from an infant skeleton from Alaska

A paper in Nature this week presents analysis of the ancient genome of an infant skeleton from Alaska, some 11,500 years old: “Terminal Pleistocene Alaskan genome reveals first founding population of Native Americans”.

Jennifer Raff has done a very nice write-up for the Guardian of the implications of these new data for models of the initial habitation of the Americas: “What the ancient DNA discovery tells us about Native American ancestry”. Of particular interest is the way that this genome helps to resolve the approximate ages of population divergences as people entered Beringia and later the southern parts of the Americas.

She did not belong to either of the two major Native American genetic groups (Southern and Northern), but was equally related to both of them. One interpretation of this result is that her ancestors must have remained in Alaska after splitting from the ancestors of Native Americans sometime around 20,000 YBP. Her genome, provides new insight into the genetic diversity present in the ancestral Beringian population. One important component of that is that it gives us new estimates of the approximate dates of key events:
36,000 YBP: The ancestors of the ancient Beringians began to separate from East Asians, but gene flow between them continues until about 25,000 YBP
25-20,000 YBP: This population experienced gene flow with the ancient North Eurasian population (to which the Mal’ta boy belonged)
20,000 YBP: The ancestors of the Upward Sun River child diverged from the ancestors of other Native Americans.
17,000-14,600 YBP: The two major clades (genetic groups) of Native Americans differentiate from one another.
While this paper doesn’t yield any tremendous surprises, it does add new details to and confirms the predictions of a hypothesis for the initial peopling of the Americas that has been the focus of much research over the past few years.

I may have more to say about this later. It is of interest that this genome, like other Paleoamerican genomes so far, lacks the evidence of a slight fraction of South or Southeast Asian ancestry that some researchers have claimed to be present in some living populations from South America.

Neo one of the top 10 science images of 2017

Really honored to have one of my photos of Neo included as one of Cosmos magazine’s “Top 10 science images of 2017”.

Neo skull frontal view

It’s the frontal view of the Neo skull.

I’ve always thought that human evolutionary science has some compelling images, and we rarely see them in top science images lists, which often feature brightly colored micrographs. There’s something exotic about beautiful images that must be explained for people to see what they represent. With a fossil, the image portrays an object that speaks for itself in some ways.

This is a fun photo because I took it with old glass, a vintage 1970 Nikon manual lens that I use for many of my laboratory photos.

Link: A hookworm history in mining

I’ve been doing a bit of reading about hookworm infection for an essay, and I happened across a piece by Rebecca Kreston from Discover’s “Body Horrors” a few years ago: “Dark Pits of Disease: Mining’s History of Hookworm”.

I wasn’t really aware of the long association of hookworm and mining, later extended to tunnel-digging.

The extent of hookworm infection was properly assessed in the mid to late 1800s. The disease was recorded throughout the United States, Australia and and much of Europe (4). It was found in the gold and silver mines of Hungary, Sicily’s sulphur mines, and in the coal mines littering Germany, the Netherlands, Belgium and France (5). It was estimated that 20 to 90% of miners in Austria suffered from anemia in the late 1800s. More than half of the men slaving away in the goldmines in the California Gold Rush in the mid-nineteenth century were thought to carry hookworm, with some populations said to be infected at a rate as high as 80%. In 1916, when the California State Board of Health investigated the prevalence of hookworm among the 1400 miners based at the Grass Valley Gold Mining district in Sierra Nevada, all but two of the men were found to be harboring worms (1).

How many students in paleoanthropology can see casts of Australopithecus afarensis?

The other day I happened back upon an old post from 2005, the first full year of the blog: “NSF and data access”. The post recounts my perspective on the problems with data access back in the first decade of this century.

That post was written a couple of years after the famous “Glasnost” article in Science. That news article, by Ann Gibbons, reviewed some scientists’ perspectives on access to fossil evidence. That would prove to be an important article in documenting a phase of history.

But as I reflected at the time, the fact that scientists cannot examine recently-discovered fossils is a comparatively minor inconvenience.

The real problem is that twenty to thirty years after many fossils are uncovered, there is no cast availability, little public data access, few financial accommodations to make such access possible. Specialists like me often find ways around these barriers. But I do not think it would be overstating the problem to suggest that perhaps half the people teaching human evolution in four-year universities have never touched a cast of a Hadar fossil. I would be delighted to be proved wrong, but I don't think I am. Our field is educating students into a world in which A. afarensis is unknown in the laboratory and poorly represented in our textbooks. I'm not talking about new specimens, here, I'm talking about fossils that were found in the mid-1970's and monographed in 1982. Nor is this problem limited to early hominids. What proportion of people teaching about the modern human origins problem do you suppose have seen a cast of any "early modern" fossil other than Skhul 5?

I’ve said this quite a lot over the years. I still believe that Au. afarensis is not known in the laboratory for most instructors of biological anthropology at universities and colleges in the U.S. Indeed, I would say that a good fraction of instructors of specialty courses in paleoanthropology have not studied casts of Au. afarensis beyond LH 2 and LH 4 from Laetoli, Tanzania.

I was especially struck in 2012, when I organized the plenary session at the AAPA meetings, at which many anthropologists saw casts of fossil hominins that they had never seen before. The University of Oregon assisted by providing a second-generation cast of the Lucy skeleton. I was amazed at how many professional anthropologists clustered around the table with this cast skeleton. Many had seen a cast of Lucy mounted in a museum, but had never had the opportunity to handle parts of it, nearly forty years after its discovery.

My post from 2005 centered around the announcement of the new data access requirements for NSF grants in the physical anthropology program. I was hopeful, and reviewed several ways in which the official policy might make a difference.

But I did think that there was a real risk that the policy would give rise to extra paperwork for grant applicants without actually requiring any real changes in the way paleoanthropologists did things.

If the new policy is to be a success, then the proof of it cannot wait for ten to thirty years. It needs teeth. It needs two or three high-profile grants to be declined because of data access issues. And it needs those cases to be made public, so that everyone can have confidence in the openness of the process. This doesn't mean that the names of the applicants and their alleged sharing violations should be dragged through the press. It does mean that NSF should publish the number of grants (and their proposed funding amounts) declined for failings in the data access plan.

There has been positive progress. The data archive at Duke University, Morphosource, initially funded by NSF, has become an important data repository for surface scan data from some recent fossil hominin discoveries. None of those hominin fossil discoveries were themselves funded under NSF grants, but NSF funding did go toward recovery of a number of Turkana Basin hominin fossil specimens, for which photogrammetric models are available under the AfricanFossils.org initiative.

A cultural change is underway in data access and accessibility of 3-D models and casts of fossil hominin specimens. I am much more hopeful about the future of data access now than I was in 2005.

With fossils like those of Australopithecus afarensis, there are very specific heritage issues that pertain to the country of origin and limit scientists’ ability to share information about their fossils.

For example, researchers who work in Ethiopia have told me that only a single cast of any fossil specimen is allowed to be exported from the country by the discoverer. Obviously that requirement severely limits information exchange about such discoveries among professional anthropologists, much less the public. It also severely reduces the broader impact of fossil discoveries, unfortunately, because of the extreme difficulty of including them in comparative analysis.

I will be reflecting on these issues much more over the coming months. We are on the verge of some real changes in paleoanthropology’s profile in science and its impact on people’s awareness of science. A critical mass of anthropologists recognize the importance of working together on issues of access and data replicability.

Lessons of Piltdown doubters

Darren Naish at Tetrapod Zoology has published a nice post about early doubters of the Piltdown fossils: “Piltdown Man and the Dualist Contention”.

The scientific establishment recognized that the supposed Piltdown hominin fossils were forgeries in the early 1950s. But long before that time, even shortly after the initial description of the fossils, many scientists doubted that the specimens actually could have came from a single individual. The jaw fragment attributed to Piltdown 1 was too much like those of living apes, and the skull fragment too humanlike for them to make much sense as skull and jaw of a single primate.

Maybe the most far-seeing of the critics was Gerrit Smith Miller:

Miller (1915) concluded that the jaw represented a European chimp which he dubbed Pan vetus. The cranium, he thought, should be referred to the genus Homo and the name Eoanthropus, he proposed, should be discarded. Because the jaw and cranium were missing those parts that would ordinarily reveal the way in which they articulated, Miller (1915) noted that “Deliberate malice would 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”.

Naish discusses Aleš Hrdlička’s reaction to the fossils. Hrdlička is remembered in American archaeology as one of the greatest skeptical thinkers in the history of science. He adopted the hypothesis that humans had reached the Americas only after glacial times, and famously worked toward debunking several claims of extremely early human skeletons in the United States. Hrdlička published his observations on the original Piltdown specimens in 1922, with several passages that deserve quoting.

Here’s what he had to say about the lack of congruity between the mandible and skull of Piltdown 1:

The first strong impression which the [mandibular] specimen conveys is that of normality, shapeliness and relative gracility of build rather than massiveness. When, after studying the specimen for a good part of two days, the observer took in had the thick Piltdown skull, there was a strong feeling of incongruity and lack of relationship, and this feeling only grew on further study. As a rule there exists a marked correlation between the massivity of the skull--particularly if as in this case the upper facial parts were involved in the same--and the lower jaw. A finely chiselled mandible of medium or sub-medium strength belongs as a rule to a skull that is characterized in the same way, and vice versa. To connect the shapely, wholly normal Piltdown jaw with the gross, heavy Piltdown skull into the same individual, seems very difficult. After prolonged handling of both the jaw and the skull there remained in the writer a strong impression that the two may not belong together, or that if they do the case is totally exceptional.

That passage may seem striking, not because of how right Hrdlička was, but of how wrong he was.

Hrdlička formed the opinion that the jaw fragment was fundamentally different from chimpanzee jaws, and therefore likely a human ancestor. That conclusion may have owed much to the fact that Hrdlička did not compare it to orangutans. At any rate, his conclusion was the opposite of that formed by Gerrit Smith Miller, who believed the jaw must have belonged to a chimpanzee, while the skull was fundamentally human. Hrdlička did not undertake a detailed examination of the skull in his 1922 paper, and seems to have thought it was massive and robust, setting it apart from recent humans.

About the molar labeled as a part of “Piltdown 2”, which Dawson had claimed was found 2 miles from the skull and mandible of Piltdown 1, Hrdlička had this to say:

The additional molar tooth of the Piltdown remains is in every respect so much like the first molar of the Piltdown jaw, that its procedure from the same jaw seems certain, and it would seem probable that the account of its having been discovered at a considerable distance away might be mistaken. The tooth agrees with those of the jaw perfectly not only in dimensions and every morphological character, but also in the degree and kind of wear. A duplication of all this in two distinct individuals would be almost impossible.

Hrdlička may not have intended this to be quite as provocative a claim as it might appear today. At the time that Arthur Smith Woodward revealed the Piltdown 2 discovery in 1917, Charles Dawson (who supposedly discovered the specimens) had been dead for a year. Hrdlička certainly was suggesting carelessness by Dawson, or miscommunication between Dawson and Smith Woodward. After all, Hrdlička had seen many instances of careless specimen collection leading to misunderstandings in the United States. With Dawson dead there was no way of simply asking.

Eleven years later, Smith Woodward replied to Hrdkička’s comment in a short letter to Nature.

Fortunately, among old correspondence, I have just found a postcard of July 30, 1915, written by Mr. Dawson, in which he announces his discovery of this molar tooth "with the new series". I have given the postcard to the Geological Department of the British Museum (Natural History), so that the record may be preserved and made available for reference.

I can’t help but read Hrdlička with the strong sense that if this had been an early human discovery in the United States, his skepticism about the fossil’s reported context would have been more potent. In the U.K., postcards from fossil collectors and the word of prestigious scientists provided all the context anyone could ask.

Many historians of science have discussed the incongruity between the ready acceptance, at least within the English scientific establishment, of the provenience and purported great geological age of the Piltdown material, and the great skepticism–often by the same scientists–of discoveries in Africa.

A number of anthropologists, led by Dart himself along with Phillip Tobias, thought that the Piltdown discovery made for a “perfect storm” of doubt about Australopithecus. By the standards of the 1910s and 1920s, the Piltdown discoveries came with almost impeccable geological context, accompanied as they were with well-known Early Pleistocene mammal bones.

This is a scientific case in which the skeptics were many, even from the start. Hindsight tells us that skepticism was fully justified. One lesson that students should learn about this case is the value of open airing and discussion of skeptical views.

Full body reconstruction of Piltdown
Full body artist's reconstruction of "Sussex Man", by Amédée Forestier in the London Illustrated News, 1912.

The hoax should have been obvious even in 1912, from file marks on the teeth, but these were overlooked. Contemporary skeptics could not openly admit, or did not dare guess, that the bones were the product of a deliberate and malicious forgery.

At the same time, the skeptics showed many of the weaknesses of the science in the early years of the twentieth century. The impressions of skeptical professionals often totally contradicted each other. They could look at the same specimen and have opposite conclusions. Hrdlička and Miller are two great examples of this. They both disbelieved the conclusions drawn by people like Smith Woodward and Arthur Keith. But at the same time, they disagreed almost fundamentally with each other. Their observations were careful, but they lacked a comparative sample. Hrdlička defends his observations by saying that Miller didn’t have the opportunity to inspect the original specimens.

Today, at the beginning of the twenty-first century, some professional anthropologists still deploy that explanation. They say that observations must be taken on original specimens, and scientific descriptions cannot encompass the details visible in originals.

To me, that is the opposite of a scientific perspective. Sure, a poor description may misrepresent an original fossil specimen. Attempts at reconstruction may fail to achieve a true representation of the anatomy.

But we must work toward improving these descriptions. If they cannot be replicated, then they cannot inform our scientific ideas.

It took new technology and the death of a generation of scientists to overturn the place of the Piltdown fossils in human evolution. Getting rare and exceptional evidence wrong caused years of mistaken effort. Paleoanthropologists must grapple with rare, sometimes singular, evidence. Other sciences deal in larger sample sizes, where unusual data points can be ignored as outliers. But in paleoanthropology, “outliers” cannot be ignored.

Yet in a field where many finds are made on land surfaces with no in situ geological context, we must also recognize that some fossils may appear to be outliers because their discoverers have erred in their context. That problem continues, even without deliberate falsehoods like the Piltdown specimen. We must find ways to strengthen context, and be clear about the cases where context is weak.


De Groote, I., Flink, L. G., Abbas, R., Bello, S. M., Burgia, L., Buck, L. T., ... & Kruszynski, R. (2016). New genetic and morphological evidence suggests a single hoaxer created ‘Piltdown man’. Royal Society Open Science, 3(8), 160328.

Hrdlička, A. (1922). The Piltdown jaw. American Journal of Physical Anthropology, 5(4), 337-347.

Woodward, A. S. (1933). The Second Piltdown Skull. Nature, 131, 242-242.

Straus, W. L. (1954). The great Piltdown hoax. Science, 119(3087), 265-269.

The New Archaeology, social laws, and playboys

Sick burn by Bruce Trigger, 1984:

By the 1950s, a growing number of archaeologists were smarting from the charge that their discipline was descriptive rather than theoretical in orientation and that they were the not very intelligent playboys of anthropology. Many ethnologists were claiming that their own work was more nomothetic in orientation than it appears to be today (99). This made many archaeologists anxious to prove that they could do whatever ethnologists could. Among this group, the New Archaeologists dedicated themselves to using archaeological data to contribute to the development of a general body of social science theory (112, pp. 364-68; 180). At the same time, they borrowed from general anthropology, and in particular from the work of Steward and White, a set of concepts that were not shared at that time by a majority of ethnologists and which remain controversial (87, pp. 117-341). These were chosen, not because they were demonstrated to be better founded than others, but because they appeared to enhance the theoretical importance of archaeological data.

This strikes me as weird from today’s perspective.

It is true that ethnologists in the 1950s aspired to “general theoretical explanation”. People like Leslie White and Julian Steward were trying to formulate general laws of social behavior, and some archaeologists aspired to do them one better by providing historical evidence of processes over time.

But I have to read this in light of the complete revolution in social anthropology of the 1980s and 1990s. Ethnology remains “theory-oriented”, but the theory that matters is highly particularized, not general. Cultural anthropologists recently have not tried to formulate social laws, they have tried to arrive at interpretive descriptions of social events in their particular cultural contexts.

In other words, cultural anthropology today is a lot like what Trigger describes as the “culture-historical” school of archaeology prior to 1940.

There are today social scientists who are trying to understand general social laws of historical change, but they mostly don’t identify themselves as anthropologists.

Archaeologists’ role in all this has changed since the 1980s, and that deserves more discussion.

A view of the Little Foot skeleton

Paleoanthropologist Ron Clarke and the University of the Witwatersrand made a big splash last week with the public unveiling of one of the most important hominin fossils ever discovered, known as “Little Foot”.

Little Foot skeleton
The Little Foot skeleton, photo: University of the Witwatersrand

The press announcement was not accompanied by new data or scientific studies on the skeleton, which were said to be forthcoming.

I’ve written some thoughts about this discovery and its scientific importance on Medium: “Will the “most complete skeleton ever” transform human origins?”

By all accounts, the Little Foot skeleton is not within our genus, Homo, nor is it confounded at the base of our genus, as Australopithecus sediba may be. Whatever its identity, whether it is Australopithecus africanus or “Australopithecus prometheus”, the Little Foot skeleton has an essential role for testing hypotheses of how species are related within Homo. For such hypotheses, Litttle Foot is what biologists call an “outgroup”, the most complete and closest one to our genus ever discovered.
Will it confirm old ideas, or overturn them? Obviously this one skeleton won’t answer every question. But now, no study will be sufficient without it.

The University of the Witwatersrand also released some videos giving background of the discovery and the press event last week, which are worth watching.

Should anyone be worried about the number of scientific research papers that are never cited?

Nature looks at the myth that a large fraction of scientific research goes uncited in a piece by Richard Van Noorden: “The science that’s never been cited”. Compiling a number of stories from researchers who have looked into citation rates, the piece concludes that around 10% of scientific research papers in journals tracked by Web of Science will go uncited, a bit more than this if self-citations are excluded. That number varies greatly by technical field, with engineering disciplines much higher.

None of these compare to the urban myth about citations, which is that half of papers go uncited. But really, I’ve heard that said about humanities papers, not scientific papers, and Van Noorden’s article acknowledges that research in the humanities tends to be more independent, with a higher fraction of research that goes uncited by other workers. The Nature piece traces to urban myth to Science:

The idea that the literature is awash with uncited research goes back to a pair of articles in Science — in the 1990 one and in another, in 1991. The 1990 report noted that 55% of articles published between 1981 and 1985 hadn’t been cited in the 5 years after their publication. But those analyses are misleading, mainly because the publications they counted included documents such as letters, corrections, meeting abstracts and other editorial material, which wouldn’t usually get cited. If these are removed, leaving only research papers and review articles, rates of uncitedness plummet. Extending the cut-off past five years reduces the rates even more.

Should we even worry about citations? After all, basic research is its own reward. The Nature article at some points reads like a support session for scientists who aren’t feeling the citation love, as it goes through one reason after another that research may be valuable even if no one ever cites it.

Most of the reasons discussed in the article are totally legitimate, and should be part of any conversation about the value of lines of research that don’t provoke lots of additional work on exactly the same model.

Still other articles might remain uncited because they close off unproductive avenues of research, says Niklaas Buurma, a chemist at Cardiff University, UK. In 2003, Buurma and colleagues published a paper about ‘the isochoric controversy’ — an argument about whether it would be useful to stop a solvent from contracting or expanding during a reaction, as usually occurs when temperatures change11. In theory, this technically challenging experiment might offer insight into how solvents influence chemical reaction rates. But Buurma’s tests showed that chemists don’t learn new information from this type of experiment. “We set out to show that something was not worth doing — and we showed it,” he says. “I am quite proud of this as a fully uncitable paper,” he adds.

A good amount of research in human evolution may fall into this category.

Case-control studies on excavation practices, for example, are very rare. It would be tremendously valuable to know whether some common variations in practice make a difference to data recovery, or whether innovations might result in better data. Failed experiments that nonetheless reinforce the value of existing practice are valuable knowledge, but probably shouldn’t need to be cited again and again.

All in all, it’s wrong to think about “landmark findings” giving rise to lots of citations. The most cited papers are those that establish new experimental (or computional) methods, and those that provide datasets useful for other researchers. Those are very good things, but not the only things!

Link: Language development in the Tsimané

A nice article in Scientific American by Dana Smith looks at a new study of language development in the Tsimané people of Bolivia: “Parents in a Remote Amazon Village Barely Talk to Their Babies—and the Kids Are Fine”.

The researchers observed, anecdotally, that language development appears to be slightly delayed in the Tsimané—but this does not seem to matter. The children grow up to be fully functioning, communicative and productive members of the community. In fact, as interactions between Tsimané and other Bolivians increase, many of the children are becoming bilingual in Spanish as well at their native Tsimané language.

This is a good story of the way that differences in childrearing across cultures have unpredictable outcomes, and what has been recommended within particular Western societies may not generalize to other places.

Link: An appreciation of Frank Brown

Nature last week published an appreciation by Bernard Wood of the life and contributions of the late Frank Brown, who died earlier this fall: “Frank Brown (1943-2017)”. The first paragraph gives a good summary of the importance of Brown’s work:

Although not a palaeoanthropologist, Frank Brown played a major part in unveiling the story of human evolution. He devoted half a century to working out the complex geology of the fossil-rich sediments of East Africa’s Omo–Turkana Basin, one of the key sources of information about early human evolution. By matching up the chemical signatures of volcanic ash layers identified at sites across the basin, Brown provided a reliable way to place fossil and archeological finds in chronological order, adding immeasurably to what we know about human origins.

Brown and his students and collaborators contributed the intellectual basis on which we now understand the Rift Valley chronology of human origins.