Human skeletal remains are not a common occurrence at forensic scenes but, when present, they can preserve a great deal of information about the events surrounding the deposition of a corpse. Forensic anthropology, the field that specializes in the analysis of human skeletal remains in medico-legal contexts, has garnered a good deal of attention from the popular media in recent years. Most of that attention comes from Fox's drama series Bones, which ran from 2005 to 2017 and followed the exploits of forensic anthropologist Temperence Brennan (a character from the novels of forensic anthropologist Kathy Reichs). Suffice it to say, forensic anthropology is now among the most popular courses offered in anthropology departments (including at UNCG where I teach), and undergraduate concentrations and graduate programs in forensic anthropology have popped up across the country.
Forensic anthropology encompasses a wide variety of methods and techniques, one of which is forensic taphonomy. Taphonomy, broadly defined, is the study of how organic remains (like skeletons) transition from living to static entities. In a forensic context, we ask the question "what happens to a person between death and discovery"? A class of taphonomic data, known as bone surface modifications, or BSMs, is particularly useful for answering this question. Marks on bone surfaces, like carnivore tooth marks, saw marks, insect damage, and the like, can provide critical evidence concerning an individual's death and their body's journey after death. Some insects, for example, are only active on the surface, so if we find damage from that particular insect on human bones, we can be sure that the bones were exposed on the surface, at least for some period of time. A few years ago, in a high-profile murder case in North Carolina, where I live, marks on the bones of a human skeleton were used to help identify the saw used to dismember a body.
Last fall, my friend and colleague, Travis Pickering, who has spent the better part of 25 years studying taphonomy, was solicited to write a review on the use of BSMs in forensic anthropology for the journal WIREs Forensic Science. He graciously invited me to collaborate on the undertaking, and he and I spent the last year researching how BSMs were integrated into the forensic sciences and how they've been used in crime scene reconstruction. The result of those efforts is an article titled "Cruel traces: bone surface modifications and their relevance to forensic science." (I wish I could take credit for the clever title, but that was all Travis.) As much work as they are, I enjoy writing review articles that cover such a wide swath of a particular field of inquiry because they force me to explore literature that I otherwise might not even have come into contact with.
Several key themes emerge in our review, namely that: (1) the study of BSMs developed first in the paleontology and archaeology of the mid- to late-1800s and was later adopted by forensic investigators; (2) ultimately, a BSM's utility for forensics (or any other field) is only as good as our ability to link that BSM with a particular process, and that link can only be established through systematic observations of a process actually producing a specific BSM; (3) the features of the BSM itself are not always enough to provide a positive identification, and contextual information must always be used (e.g., serrated knives and shark teeth create very similar striations on bones surfaces, but if the bones were recovered far inland, a shark origin is much less likely); (4) the admissibility of expert scientific testimony in federal courts (the so-called "Daubert" standards) privilege the validity of a scientific methodology over the expertise and experience of any one scientific expert.
As an example, consider a linear striation on the surface of a bone recovered from a crime scene. Now, we might logically interpret that mark to have been produced by a knife. That's all well and good, but lots of things can create a linear mark on a bone: a small, sharp sand grain on the ground, the tooth of a scavenging dog, and so on. How might we know that the knife is, in fact, the source of the mark? Well, for one, we need to conduct an experiment where a knife is observed to create a particular mark on a bone's surface (this can be done with donated human skeletal material or, more commonly, the bones of other mammals like deer or pigs). That way, we know for sure what a knife mark looks like. There are lots of different types of marks, though, and even the same knife can create slightly different marks depending on how it is wielded, so one experiment is typically not enough. However, what if a knife was found alongside the human skeletal material? This sort of contextual information, when considered along with the morphology of the mark itself, can provide additional clues and make our identification that much more probable. When this evidence is presented in court, its admissibility, at least according to federal standards, will have less to do with the experience of the expert witness than with the rigor of the methodology used to identify the BSM.
Importantly, no methodology can definitively, with 100% confidence, identify the source of a particular BSM. Why not? Well, forensic investigators do not themselves witness a crime: the relevant details must be reconstructed. This is where we get the phrase "crime scene reconstruction" and its common association with archaeology, and for good reason. Forensic investigators are in many ways like archaeologists: whereas archaeologists use artifacts and architecture to reconstruct ancient cultures, forensic investigators use evidence to reconstruct a crime. This means, though, that we must work with probabilities, rather than absolute certainties. When we are able to match up a BSM from a crime scene to an experimentally produced BSM, it becomes highly probable, thought not absolutely certain, that the object that created the crime BSM is the same as the object that created the experimental BSM.
A major hurdle for forensic investigators is that some BSMs can be difficult to tell apart from each other. The marks produced by different types of saws, for instance, share many features. So, too, do the marks produced by different types of knives. With so much overlap, how similar do the little bumps, grooves, and edges of a crime BSM have to be to an experimental BSM to be considered "the same"? What features should we even be looking at, and how do we define them? One analyst's "deep groove" may be another's "V-shaped striation." This sort of inconsistency makes it difficult for BSM analysis to attain Daubert-level methodological rigor. Travis and I conclude with the suggestion that computerized image analysis of BSMs coupled with statistical classification might get us closer to reaching that level of rigor. Either way, it will be interesting to see how the field progresses from here.
