Caring for the edentulous

One of the features of the National Geographic (April 2005) article on Dmanisi is the discussion of the necessity of other people to aid and care for the old and infirm. The skull D3444 is an old adult individual, with no teeth remaining and the alveolar surfaces of the jaws nearly completely resorbed. This means that the individual lived without functional teeth for a period of time extending well beyond a year, and possibly to several years.

The article quotes some of the researchers working on the fossils about the importance of the condition of this individual:

In the survival of the old man, "we're looking at perhaps the first sign of truly human behavior in one of our ancestors," says [David] Lordkipanidze. It could be a glimpse of a new level of planning and sharing, adds Philip Rightmire, an anthropologist at Binghamton University in New York State who is one the Dmanisi research team. "Seeing this at the very dawn of Homo, our own genus," he says, "may be the most exciting thing of all." (Fischman 2005:19)

The idea is that a person who could not chew could not survive on his or her own. So he or she (I'm sticking to "he or she" until I see more information about D3444; it is more robust than the two females, but not extremely so) needed help from other people, in this story. The interesting aspect of this specimen is that it is so much older than the next oldest specimens with comparably extensive tooth loss, which are all Neandertals. Dmanisi does appear to considerably extend the time period over which we have evidence for human survival in the face of extensive disability; at least in terms of dental function.

There has been some criticism of the idea that tooth loss is a necessary indicator of care from other individuals.

The observation of old, edentulous individuals in a number of species of primates has formed the basis of a disagreement about the importance of such individuals for inferring the behavioral capacities of early hominids. This literature is reviewed by Cuozzo and Sauther (2004), and includes two examinations of the Aubesier 11 Neandertal by Lebel and Trinkaus and two critiques by David DeGusta. The gist of the argument is a discussion of whether any wild primates are observed to be missing as many teeth as certain Neandertal specimens (particularly Aubesier 11, but one may also include Monte Circeo, La Chapelle-aux-Saints, and possibly others who retain a higher proportion than Aubesier 11).

In my view, the actual proportion of missing teeth is much less important than the overall view of the function of the dentition. A chimpanzee with no functional occlusion is certainly almost as poorly off as a Neandertal with no teeth at all, even if many teeth are still present.

As an example, here is a shot of a female chimpanzee palate in the CMNH collection:

Chimpanzee palate with extensive premortem tooth loss
Chimpanzee palate showing extensive premortem tooth loss. Photo: John Hawks.

This is one out of around fifty wild-shot chimpanzees, a collection that is biased toward younger individuals than would have died natural deaths. The mandible retains the premolars and molars, although only one incisor, so the individual had essentially no functional dental occlusion. All chewing capacity was tooth against gum. As such, this chimpanzee wasn't quite as poorly off as the D3444 individual, but was certainly comparable in terms of dental dysfunction. There are several chimpanzee crania in this collection that are missing a few teeth, although none as extensive as this one (1).

But I don't think this line of argument is particularly productive, because it evades the central issue: do rare individuals ever allow any inferences about the social attributes of ancient humans? The edentulous hominids are evidence of the extreme end of a range of variation in life history traits. DeGusta (2002, 2003) essentially argues that the extreme end for Middle Pleistocene humans is not greatly different, if at all, from that in other primate species. I am sympathetic with that view, but I don't think it goes far enough to answer the problem. For one thing, the data as they stand indicate that Neandertals actually did survive with worse health status than other primates. This is not only true of tooth loss (where the difference is quite minor) but also bone breakage, arthritis, cranial injuries, and other assessments of both trauma and chronic health conditions. For earlier humans, the data are sparser, but these people had their problems also, as evidenced by D3444 and specimens like KNM-ER 1808. I suspect that the sample of Early Pleistocene hominids as it stands is not significantly different from Neandertals in health (although it is significantly different in longevity). So we cannot let the matter rest on the idea that other primates are like early humans in end-of-life health status; it is quite likely that early humans were not very much like any other primate.

DeGusta (2003) includes another objection to the assumption that edentulous individuals indicate care from other individuals:

Lebel and Trinkaus (2003) and Lebel et al. (2001) fail to suggest any reason why Aubesier 11 would have been unable to obtain or manually process soft foods on his/her own, rather than relying on conspecifics to do so. There is no evidence of any condition, or even advanced age, that would have precluded Aubesier 11 from doing so. So even granting the rest of their argument, Aubesier 11 cannot be used as evidence of conspecific care (92).

To take the position that these individuals are strong evidence of social behavior is to make several assumptions:

  1. That a change in the average pattern of behavior highly affects the extreme end of the range of life history.
  2. That it is social interaction and not some other behavioral or life history change (e.g. diet, disease, day range, secondary altriciality) that is responsible for the difference.
  3. That the end of the range is sampled adequately to make such inferences.

I think all these assumptions are unwarranted.

In the case of life history variation, I think that the survival of a small number of individuals under extraordinary circumstances says little about the habitual capabilities of a species. The difference between extraordinary and ordinary is one of sampling density. One individual out of a sample of five may be entirely normal, or a one-in-hundreds freak occurrence. Ten individuals out of fifty, while the same proportion of a sample, clearly are not exceptional.

In any population, some individuals are likely to survive under circumstances that would usually be fatal. For example, food availability varies greatly both from place to place and from year to year. Although the odds of mortality are higher for older adults, in practical terms these odds fluctuate along with ecological conditions. There are likely to be periods of years when a very low proportion of older individuals die, and some survive with infirmities that are extreme for their population. Older edentulous individuals who are otherwise healthy have a number of advantages. These range from a relatively high social status (and thereby a claim on food noticed by others) to extensive knowledge of food sources and other ecological needs, to a greater ability to evade or resist predators.

We can ask a more basic question. Is antemortem tooth loss in humans evidence of aid from other individuals? A high proportion of older adult humans today lack functional dentitions. In industrialized societies and to a great extent elsewhere, these people make use of artificial dentitions, which we can presume were not part of the technological repertoire of Early Pleistocene humans. But dentures and other dental appliances are unavailable to a substantial proportion of edentulous humans today, for economic and other cultural reasons. These people do not starve; instead they use extensive extraoral processing to enable the consumption of a relatively normal diet. This does not require, although it sometimes involves, the assistance of other people.

Certainly the present-day situation is different from that experienced by the Dmanisi hominids, or even Neandertals for that matter. Edentulous people today are greatly aided by the consumption of a high-starch diet of grains, tubers, or other stored vegetables that can be reduced with long cooking to a paste or mush. But among living people such a diet is routine, and long survival after the loss of a functional dentition is very common. In Pleistocene humans, such survival was almost certainly exceptional, as argued by the low proportion of edentulous remains.

It is no great stretch to think that an occasional older person might have put together a diet for several years that would allow survival without teeth. Such a diet need not have been steady or nutritionally complete. We can imagine the life of such a person, possibly with long stretches of hunger punctuated by a rare full meal on soft plant foods, honey, organ meat from an animal, or chunks of flesh painstakingly sliced thin with a stone flake. We do not even need to imagine that such foods would have been cooked, as supported by the survival of edentulous, non-cooking primates.

Tooth loss and selection

There is another, possibly more interesting, question arising from this specimen. Presumably the dentition is adapted to the life history of a species. Long-lived species have teeth that last a long time; short-lived species need not have teeth that last as long. Under some circumstances, there is value to having teeth that have thinner enamel (and therefore wear more rapidly), are smaller (and therefore wear more rapidly), or otherwise do not last as long. Smaller teeth may allow the application of greater masticatory force to certain kinds of food items (such as pliable plant or animal muscle fibers). Thinner enamel allows an enamel/dentine wear gradient that maintains greater occlusal topography for more effective shearing of food. But such teeth are not well-suited to a long lifespan unless the rate of attrition can be reduced by diet choice. These contrary influences on tooth form lead to different stable equilibria in different species, depending upon their life history and diet.

The interesting question is the strength of selection resulting from loss of dental function in old individuals. These people survive for some length of time with compromised teeth. Although the causes of tooth loss and extreme dental wear are not always the same, both factors lead to a reduction in dental function. This is especially true for old people who have had lives leading to high degrees of attrition or dental disease; which may occur more or less depending on the prevailing environmental conditions, social status of the individual, and possibly dietary preferences.


1. As an interesting aside, I scored dental wear in over 100 male gorillas from this collection, and I do not remember that any of them had any significant number of teeth missing. This recollection accords with the data presented by Cuozzo and Sauther (2004), where they find no gorillas out of 65 that have more than 40 percent of their teeth missing. [UPDATE 4/26/05: A reader points out that these observations come from Nancy Lovell's (1990) work, cited by Cuozzo and Sauther (2004).] To be honest, I don't remember there being a single specimen missing that many teeth, although I deliberately excluded specimens with missing teeth from my own wear sample so I surely don't remember them as well as the ones I used.


Cuozzo FP, Sauther ML. 2004. Tooth loss, survival, and resource use in wild ring-tailed lemurs (Lemur catta): Implications for inferring conspecific care in fossil hominids. J Hum Evol 46:623-631.

DeGusta D. 2002. Comparative skeletal pathology and the case for conspecific care in middle Pleistocene hominids. J Archaeol Sci 29:1435-1438.

DeGusta D. 2003. Aubesier 11 is not evidence of Neanderthal conspecific care. J Hum Evol 45:91-94.

Fischman J. 2005. Family ties: Dmanisi find. National Geographic April, 2005:17-27.

Lovell NC. 1990. Patterns of Injury and Illness in Great Apes: A Skeletal Analysis. Smithsonian Institution Press, Washington DC.

Lebel S, Trinkaus E. 2002. Middle Pleistocene human remains from the Bau de l'Aubesier. J Hum Evol 43:659-685.

Lebel S, Trinkaus E, Faure M, Fernandez P, Guérin C, Richter D, Mercier N, Valladas H, Wagner G. 2001. Comparative morphology and paleobiology of middle Pleistocene human remains from the Bau de l'Aubesier, Vaucluse, France. Proc Natl Acad Sci U S A 98:11097-11102.