Place, Place Baby



More than sciences like physics, which can replicate experiments in laboratories, paleontology is a science that involves rare, fragile objects that are situated in particular locations—making the practices shaped by those locations a crucial aspect of the discipline. (Madison 2019, 22)

There is a tension built into science: a tension between data’s local specificity, and the more global aspirations scientists have for their claims. Data is generated in particular times and places, using particular methods, often beholden to peculiar circumstances. But then data must ‘travel’—it must depart from the idiosyncrasy of its origins to be integrated with other data and used to construct evidential claims.

Paige Madison’s recent paper “All things bleak and bare beneath a brazen sky: practice and place in the analysis of Australopithecus” examines a case which represents this tension well. Her story follows the ‘Taungs Baby’—the first Australopithecus (of the same genus as the more famous Lucy, discovered in Ethiopia in 1974). Uncovered in 1924, the Taungs Baby became the centre of a scientific controversy, with the Australian-born, London-educated, and South African-resident scientist Raymond Dart arguing that he had found an early ancestor of our own species (Dart 1925). During the 5 years after the specimen’s discovery—which Madison focuses on—Dart’s claims were dismissed. Only much later was the Taungs Baby accepted as a critically important find. Madison elegantly situates her case study within the imperial politics and scientific mores of science during the 1920s.

A cast of the fossil skull known as the ‘Taungs Baby,” or “Taung Child,” shot from two different angles. Images from Wikimedia Commons.

Madison is concerned with explaining the initial dismissal of Dart’s claim. Previous historical treatments focus on either the personalities of the various scientists involved, or on what Madison calls ‘theoretical’ issues: the notion that humans didn’t evolve in Africa, that the brain was the first human trait to have evolved, and so forth. And indeed, these play their roles in the hypothesis’ initial rejection. But Madison thinks these accounts leave out the story’s main driver:

Evaluating the similarities and differences of characteristics across specimens has long been considered necessary for scientists to understand the range of biological variation, organize specimens, and understand evolutionary relationships. Knowledge is therefore derived from—and dependent on—collections of objects, places where scientists could measure variations, compare features, and create knowledge. (Madison 2019, 10; references removed)

According to the main contending alternative, Dart had found the skull of a juvenile ape. To establish that his specimen was a species of hominin, then, Dart had to carefully compare the Taungs Baby to other specimens. And this required access to those specimens. Dart was at the relative fringe of Britain’s scientific empire, in South Africa, thousands of miles from the great collections of Europe. Thus, he was unable to make the comparisons directly. Madison points out that science during the imperial period involved a centralizing process as specimens from all over the world were collected/stolen and sent to museums in Europe. These museums and other collections served as what Bruno Latour (1987) has called ‘centres of accumulation’. They provided organized locations for the storage and study of vast numbers of specimens while research traditions and norms grew up around them. 

So, why wasn’t the South African specimen sent to one such centre? According to Madison, this is due to changing colonial attitudes in the early twentieth century. Some colonies and dominions were asserting their autonomy from centralizing forces, and important discoveries started to be seen as objects of local pride; something making that place unique, rather than fodder for the glory of global empire. No wonder then that Dart and his colleagues were unwilling to give up the specimen. Further, practitioners in South Africa (unlike those in London) lacked the resources and know-how required to produce accurate casts, forcing Dart to rely on sculptures of the model. And, as Lukas Rieppel has argued here at Extinct and elsewhere (e.g., Rieppel 2015), there’s a big epistemic difference between a cast and a sculpture. So, unwilling to give up his fossil, and unable to produce replicas which would enable others to verify his findings in ‘centres of accumulation’, Dart’s claims were unable to meet the standards of the day. Thus, they were rejected.

Madison doesn’t present this explanation as being in conflict with others—it is rather complementary.  

While these factors—theory, ego, or perceived authority—certainly played some role in the debates, I have shown that the concerns of these “cliquish” scientists were grounded in deeply embedded historical practice of circulating a hominid fossil to a scientific center. (Madison 2019, 20)

And this particular case, Madison suggests, could act as a template for a research program. Historians interested in hominid fossils should be “asking how the trajectories and receptions of the specimens differed, how these practices have shifted over time, and how they were shaped by individual factors of location, resources and time period” (Madison 2019, 21). Madison, then, accounts for the early fate of Dart’s claims in terms of the norms and practices of early twentieth-century science, its relationship with increasing assertions of autonomy in European colonies, and the material nature of paleontological evidence.

Adrian writes…

Paige Madison’s (2019) paper hits two beats that really interest me. First, the importance of comparative work in the life sciences. The centralized fossil-hubs which Dart so lacked mattered epistemically because analysing a fossil, and particularly situating said critter in a phylogenetic context (that is, inferring who its relatives are) requires careful comparison between its morphological structure and those of its putative relatives. Emphasizing this comparative context, I think, goes a long way towards explaining how paleontologists can manage big inferences from apparently tiny data sets (from a single tooth to an extinct platypus, for instance…). Second, Madison’s major themes are the importance of materiality and institutional structures—the nature of data—for historical science. The fact that some fossils are great big hunks of funny-shaped rock make a profound difference to knowledge-making in vertebrate paleontology. These things need to be stored, organized, and prepared before they may be deployed as evidence. And although paleontology has changed a lot in the century-ish since the Taungs Baby’s discovery, it hasn’t changed that much. Much paleontological argument (including the recent challenge to the traditional ‘bird-hipped’ versus ‘lizard-hipped’ phylogenetic split between dinosaurs) turn just as much on explicit morphological analyses and re-analyses as they do on fancy computational number-crunching. And moreover, people often massively underestimate the difficulty of digitization: the resources required to scan large numbers of fossils, let alone figure out how to store and organize them, are not so easily come by. So, the materiality and location of specimens matters for vertebrate paleontology these days too.

Madison’s historical work highlights important features of paleontological practice and epistemology that are close to my philosophical heart. Here, I want to reflect on the distinction between ‘theory’ and ‘practice’ implied in Madison’s argument, as I think it is a little quick.

Madison’s focus on systems of practice, on how the paleontologists of the 1920s went about doing their science, is compelling. I agree that understanding practice is critically important for understanding epistemology. But I don’t think the right way of understanding practice is in opposition to theory. There are two reasons for this: first, there are theoretical practices and second, practices rely on theory. By the former, I just mean that one of the things scientists do is theorize, and that theorizing is just as much a part of a system of practice as more practical-looking-things like organizing and storing fossils. Splitting ‘practice’ from ‘theory’ feels a bit like splitting the mind from the body. Just as my cognitive activities are not easily separated from my body and the environment in which it is situated, so also is theorizing not separate from other things scientists do.

This doesn’t amount to a criticism per se—Madison’s claim is simply that appeal to particular theories (such as the idea that hominin brains evolved first) or particular attitudes (the ‘paleopolitics’ of the early twentieth century) is insufficient to explain the reception of Dart’s hypothesis. We must also examine what was required for claims being accepted at the time, and why Dart wasn’t in a position to meet those standards.

But this is where the second worry, and I think a slightly more pressing one, arises: the practices Madison points to themselves rely upon theory. These are not high-falutin’ theories about the nature of hominin evolution, but theories about under what conditions paleontological claims may be asserted—in particular, theories about the proper access, storage, and treatment specimen should undergo in order to underwrite claims. Sabina Leonelli (2016) has highlighted theories concerning best practice and the management of data-journeys as critically important for understanding ‘big data’ science. Although in Leonelli’s case the theories are somewhat more explicit, describing practices which “formalize knowledge that is taken to be widely assumed yet is usually dispersed across publications and research groups” (Leonelli 2016, 135) nonetheless, “the question is about how data are being systematized and assembled to yield understanding and what are the key conceptual ingredients and assumptions in that process” (Leonelli 2016, 136). That is to say, we should recognize how theories about data management and dissemination are a part of scientific practice. The explanation Madison proffers is highly theoretical after all, it is just that the theories in question are about the management and use of data.

Joyce writes…

The title of Paige Madison’s (2019) paper relates the “bleak and bare” academic environment Dart and company worked in to the bleakness and bareness of the ecological environment they worked in—the South African veld.  I spent time in South Africa as a child, and made several trips through the relevant ecological region (of highveld tapering into savanna bushveld).

A view of the veld at Willem Pretorius Nature Reserve in June 1996, during winter in South Africa. Photo by Cynthia Havstad.

There’s beauty there, and life too—the veld is lush in its own way, if you can see it.  Although Madison’s paper is about places, getting the titular metaphor to work involves re-inscribing some rather imperialist characterizations of the “uninviting countryside” (Madison 2019, 4)—and this kind of characterization is all that is required for her purposes—but I wish she had gone further, interrogating those characterizations and offering new ones, usurping rather than re-using familiar and, dare I say, tired stereotypes of African locales.  That’s my aim here.

Madison makes use of colonialist description of this portion of the South African countryside, relying on a 1925 article from a Johannesburg newspaper, The Star:

The landscape of Bechuanaland was desolate, characterized by open, uncultivated country known as the “veld,” a place where only a few small plants able to “survive by some miracle in the arid soil; under the scorching glare of the African sun” (“South Africa Believed” 1925).  It was an entirely unimpressive location for the discovery of an important fossil, one journalist mused, containing nothing but “bare veld, lime quarries, and a few corrugated iron sheds: all things bleak and bare beneath a brazen sky” (“South Africa Believed” 1925). (Madison 2019, 4)

But veld is not barren terrain.  ‘Veld’ is a term that refers to land that is covered in natural as opposed to cultivated vegetation.  Particular veld ecology is often characterized by its elevation and / or the particular kind of vegetation it supports.  The veld of the Taung region is no exception: it is only semi-arid (not a desert), and there are more than “only a few small plants” there.  As poetic as the quadruple alliteration of “bleak and bare beneath a brazen sky” is, I don’t think it is an apt description of this fascinating landscape—one that is characterized by a unique mix of tufa geology, savanna climate, and bushveld ecology.

On the left, the Taung skull fossil discovery site; and on the right, a shot of the surrounding landscape. Both images from EcoAfrica’s (2015) report, “Visual Impact Assessment for Improvement of Visitor Facilities, Site Infrastructure and Heritage Conservation Measures at the Taung Skull World Heritage Site,” available here.

I think that to describe this landscape as “bleak and bare” is to privilege outdated assessments, expectations, and preferences of the region’s European settlers.  The landscape does not have to be either viewed or characterized as desolate, and uninviting.  From an available, alternative viewpoint it is a stunningly beautiful area with a rich ecology.  It is far from bare, or bleak, and the sun is no more brazen in Africa than it is elsewhere (despite being endlessly described as such).

Madison’s article clearly and unequivocally establishes a dearth of crucial scientific resources in the region in 1925: an absence of easy access to what Dart needed to make his scientific case about early hominin evolution.  He lacked comparative hominid specimens, reliable shipping and transport infrastructure, access to artisans capable of creating adequate casts, and more.  But the fact that Dart found himself in a desolate academic situation does not mean that either he or the skull were stuck in a desolate ecological situation.  The fossil was indeed discovered in a limestone quarry—at the time, quite plausibly a bleak and bare place with sparse vegetation—but that desolation is an outcome of settler use and abuse of the land; it is not an ecologically characteristic state of the land itself.  We mustn’t accept, at face value, the dated descriptions of those African regions which settlers found inhospitable.  And we should not re-inscribe such descriptions, we should re-place them.


Dart, R. 1925. Australopithecus africanus: The Man-Ape of South Africa. Nature 2884(115): 195–199.

Latour, B. 1987. Science in action: How to follow scientists and engineers through society (Cambridge, MA: Harvard University Press).

Leonelli, S. 2016. Data-centric biology: A philosophical study (Chicago: Chicago University Press).

Madison, P. 2019. All things bleak and bare beneath a brazen sky: practice and place in the analysis of Australopithecus. History and Philosophy of the Life Sciences 41: 19/1–25.

Rieppel, L. 2015. Plaster-cast publishing in nineteenth-century paleontology. History of Science 53(4): 456–491.