There is an inherent problem to the justice world that has been observed in almost every common law jurisdiction. The reliability of scientific evidence in the Courtroom is at stake as well as the general functioning anchored to the presence of a decisional jury in a trial. This paper will present those issues and mainly focus its critical analysis on Forensic Anthropology, a field that is not immune from being considered as a ‘junk science’.
The concept known as ‘junk science’, first introduced by Peter W. Huber in 1991 (Huber, 1991), reflects a prominent problem that still requires attention from every science from which probative information is subject to a forensic usage (Black & Nic Daeid, 2015). Hubert once succeeded to shed light on a generalised taboo regarding issues related to scientific evidence brought to the Courtroom (Hubert, 1992). However, no solutions proposed at the time are sufficient enough to temper the controversy arisen by his publication about the admissibility of the scientific opinion of expert witnesses (Bernstein, 1996; Milich, 1994). Since then, many scholars have had their personal input on the matter that is now considered to be much more complex than an ephemeral problematic.
On the first hand, ‘junk science’ is usually used in reference to forensic science and the problem with the reliability of scientific evidence that is sometimes considered ‘[…] being stretched well beyond its limitations’ (Bernstein, 1996). Indeed, validated scientific methodology and standardisation in regard of qualifications, examination techniques and witness testimony are sometimes highly forsworn, thus leading to unreliable, non-prejudicial or simply irrelevant evidence (Christensen & Crowder, 2009; Bernstein, 1996). This first part of the problem galvanised scholars to develop a jurisprudence in order to create general guidelines concerning the admissibility of scientific evidence. In 1923, Frye v United States’ case stimulated the development of the ‘general acceptance test’. The objective was to ensure that scientific evidence would be based on the general approval of a theory from the scientific community but relying mainly on the qualification of the expert witness and the presumed veracity of science (Bernstein, 1996; Milich, 1994). This test was flawed and eventually legally dismantled for its purpose was not complete; a professional with the highest of qualifications could be conducting the examination incorrectly. In other words, valid qualifications and the contextual approval of a theory cannot guarantee the relevance as well as a non-abusive behaviour towards scientific evidence by all means (Bernstein, 1996). Years later, Daubert’s case (Daubert v. Merrell Dow Pharmaceuticals, Inc. ) led to the establishment of the ‘Reliability test’ which objective was to determinate whether or not the expert’s testimony relies upon proper scientific methods and sound reasoning (Christensen & Crowder, 2009). Nowadays, this test is still applied in accordance with four criteria in relation to the theory on which an evidence is based: (1) falsifiability, (2) peer-reviewed publication, (3) general acceptance from the scientific community as well as (4) standardisation (taking into consideration that the potential rate of error must be known and controlled) (Davies, 2005). In addition, Rule 702 had been created stating that ‘[…] trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable’ (NAS, 2009, p. 90). The state of forensic sciences was then shaken to its core as expert testimonies could no longer propose scientific evidence that is not consistent if repeated. As obvious at it may seem, the majority of forensic sciences had to undergo some serious changes in the way they produced scientific evidence in order to make sure that their methods would use verified principles and methodologies (Christensen & Crowder, 2009). The being said, the majority of common law legal systems around the world do not have the same amount of tort litigation handled by the U.S. Supreme Court but that did not lessen their necessity for a similar reform as there is a worldwide need for ‘junk science’ to be governed (NAS, 2009; Bernstein, 1996).
On the other hand, the problem also has facets that actually reside in the ability of the jury to differentiate credible scientific evidence from what can be considered as ‘junk science’. In fact, the jury is usually composed of randomly chosen uninitiated individuals who do not necessarily have a scientific background allowing them to elaborate a pertinent judgment on the evidence presented to the Court (Milich, 1994). Despite the fact that the conclusions are non-definitive, a study has shown that jurors are most of the time unable to understand complex scientific evidence being presented to them. Thus, they usually ignore the forensic opinion given by the expert witness in their decision-making process (Bernstein, 1996). Nevertheless, according to many scholars, the problem should be assessed beforehand as a matter of filtering rather than scrutinising (Christensen & Crowder, 2009; Bernstein, 1996; Faigman, 1992). That being said, this would still require considerable scientific knowledge and attribute a lot of importance to whomever (usually the judge) would be in charge of screening this information (Davies, 2005). Consequently, this leads us back to the first part of the problem; the upstream importance of solving the reliability issue of scientific evidence. However, everything cannot be solved by the desired objective of both valid and reliable scientific evidence. Where there is a jury, there is a need for vulgarisation and good communication skills free of any semantic shifts or general ambiguity. In fact, ‘junk science’ is not commonly used to describe the ability, or lack thereof, of an expert witness to communicate the evidence. However, since this part of their duty is crucial for an expert testimony to be pertinent, the numerous issues behind unreliable scientific evidence ought to recognise this as a recurrent problem. Indeed, expert witnesses also need to improve and modify the way they democratise their forensic knowledge in order to allow the jury to actually understand and work with probative data (Christensen & Crowder, 2009).
Forensic anthropology, like other forensic sciences, is now facing the aftermath of the judicial repercussions of this reform. On one side, expert witnesses are legally and ethically bound to communicate the justifications behind their conclusions (Hackman et al., 2018). On the other side, Kumho (Kumho Tire Co v. Carmichael ) or the most frequently used Daubert’s standards are now encouraging forensic practitioners not only to disclose but also prove that their scientific evidence meets the new requirements demanded by the law. However, it should be emphasised that none of those rules actually determinate nor oblige the scientists on how to conduct studies on which the evidence is based (Christensen & Crowder, 2009). That being said, they are several points that should be assessed in order for the anthropological practice to be discerned from being a ‘junk science’.
As aforementioned, forensic anthropology is not excluded from being considered as a potential ‘junk science’ as it is the final decision from the Court to decide whether are not the admissibility standards are met and accepted (Black & Nic Daeid, 2015). First of all, current methods and techniques in this field are sometimes considered too subjective to possibly meet the empirical requirements for a scientific evidence to be reliable and valid. Indeed, identification of the biological profile based on anatomical traits observable on skeletal remains is obviously associated with a certain subjectivity leading to an inter/intra-observer variation of the results. In fact, it has been proven that non-metric methods relying on experience and visual information pertaining to senescence are highly subject to inter-observer variation and that accuracy of valid data is then hard to maintain (Kimmerle et al., 2008). As a matter of fact, even metric methods are subject to a certain degree of error since osteometric measurements are sometimes unclear or simply variable; it is practically impossible to ensure that every practitioner will take the exact same measurement (Klales et al., 2012; Buikstra & Ubelaker, 1994). In addition, metric and non-metric methods based on senescence are initially subject to a certain variation as the ageing process varies a lot between individuals (Godde & Hens, 2014).
In regard of the expert opinion, subjectivity is eminently problematic as it means that the interpretation of results is open to different perspectives. Elaborating an opinion in this context could lead to a scientific war in between the prosecution and the defence rather than advising the jurors for their decision (Bernstein, 1996). One could then wonder if the real purpose of giving a scientific opinion in Court is fulfilled. On the other side, objectivity assures a certain unilateral comprehension and interpretation of the results of a particular forensic examination. In fact, objective methods can generally be ‘[…] performed by either an automated system or human examiners exercising little or no judgment’ (PCAST, 2016, p. 5) meaning that the conclusions are usually unambiguous and therefore reliable and valid for a forensic usage (Hubert, 1992). Thus, some authors believe that forensic anthropology is currently at a professional cul-de-sac, forcefully trying to modify qualitative methodologies to produce quantitative data (Christensen & Crowder, 2009). The problem persists as those modifications are incorrectly incorporated into the scientific method and because even the most relevant quantitative information is affected by those undertaking them in the first place (Dror, 2016).
Furthermore, most of the methods used in forensic anthropology are usually population specific, meaning that they have been developed based on a local geographic population (Kimmerle et al., 2008; İşcan, 2005). Also, most of them are based on archaeological individuals presenting anatomical traits influenced by taphonomy, palaeopathological changes (congenital anomalies, malformations, deficiencies, etc.) and diachronic variation of human morphology (İşcan, 2005). Then, these methods generally do not take current medical care into consideration as well as the difference of lifestyle influencing modern individuals examined in a medico-legal context when applying archaeological-based analyses (Kimmerle et al., 2008). Unfortunately, osteologists and forensic anthropologists all agree that the highest reliability is mainly obtained when seriation and comparison of human skeletal remains from the same population are possible. However, in a forensic context, human skeletal remains are usually found in isolation and/or in a state for which identification of both the individual’s biological profile or its specific ‘population’ is unknown (Walrath, 2004).
Moreover, subjectivity is usually related and illustrated by the undeniable cognitive bias of the human brain (Dror, 2016). In the United Kingdom, the Criminal Procedures Rules (2015) states that the expert duty to the Court should be to give an opinion that is both ‘[…] objective and unbiased…’ (The Stationery Office Limited, 2015, p. 149). In the United States, the most recent governmental forensic report (2016) explicitly notes the relevance of using objective methods and even states that subjective methods should be, when possible, evolved or simply replaced by objective approaches (PCAST, 2016). With the cognitive bias affecting the judgment of professionals undertaking subjective examinations, forensic anthropology seems to be no longer meeting the juridical requirements to base its scientific evidence on its current non-metric techniques. Conscious or unconscious cognitive bias can also lead to inaccurate judgment and partiality when extraneous information is being disclosed beforehand to the practitioner. Indeed, a study has shown with great causality that conclusions from forensic anthropologists to whom background information or other evidence have been given are most likely to be in favour of the disclose information (Nakhaeizadeh et al., 2014). Irrelevant contextual information communicated by police officers or other colleagues is in fact very common in the field and is thought to be almost systematic in the forensic environment (Nakhaeizadeh et al., 2014). This indubitably leads to weakness of expert testimonies and could eventually handicap the way forensic anthropology is perceived if a standard reform is not brought to the juridical field. At this point, it should not only be the responsibility of expert witnesses but rather a collective work from the whole forensic team in order to make sure to prevent those bias.
Additionally, subjectivity undoubtedly leads to systematic errors as the human brain is wired to look for similarities and ignore implicit differences, to unconsciously filter information and to rush over multiple data when a limit is reached in processing information (Nakhaeizadeh et al., 2014; Dror, 2010). Forensic anthropologists are no exception as they consistently work with a material that is characterised by a continuous variation. The conclusions based on such material can be quite hard to define from a binary point of view as they are a myriad of possibilities (Christensen & Crowder, 2009). Thus, there is an urgent need for this field as well as the other forensic sciences for studies to be done to cross-examine, precise or even create new objective methods allowing a wider array of scientific knowledge that is out of the traditional forensic mold for which reliability is a stake (Black & Nic Daeid, 2015; Dror, 2010). For example, forensic anthropologists and medical examiners have looked and interpreted butterfly fractures incorrectly for a long time before new studies actually showed how those alterations form under mechanical loads (Christensen & Crowder, 2009). This particular example shows how important it is not to take for granted a certain knowledge thought to be valid but also that qualification, experience or notoriety cannot be a guarantee of veracity.
Finally, forensic institutions around the world currently have their own standards and conduct their examinations following their own guidelines (PCAST, 2016; Christensen & Crowder, 2009). Despite the fact that they are several boards trying to assure a certain standardisation in the practice, nothing obliges a forensic anthropologist to be a member or even to rely on those institutions’ documentation as being part of his/her duties. A lack of standardisation then results in a weakness in the general practice, especially when speaking about the reliability of scientific evidence based on different standards (Daubert or even the Rule 702 for instance) (PCAST, 2016). It has to be noted that even those institutions that could lead to a potential reconciliation between the anthropological practice and the Courtroom’s new standards, are not that effective nor do they facilitate the admission of new professional members. For example, the American Board of Forensic Anthropology (ABFA) does not offer any particular mentored training (like almost every other professional board) and do not divulge any information (for example, old examinations and their entire content are kept private) to candidates before they sit for the examination (Gruspier, 2008). Instead, only general information is displayed on the website and every practitioner needs to self-train and literally hope that they have a sufficient academic knowledge to pass the board examination.
Nevertheless, forensic anthropology is certainly not the only forensic practice for which there is an urgent need for a drastic return to the basics of expert testimonies based on logic, robust and transparent scientific studies. Despite the fact that Daubert’s jurisprudence and other new standards rang the alarm for forensic evidence’s reliability, forensic anthropology is not on the edge of disappearing from the medico-legal field. Changes are slowly being brought to the fore as different associations like the Scientific Working Group for Forensic Anthropology (SWGANTH) are currently working together to develop standard guidelines that should eventually be adopted worldwide. They also meet on an annual basis to highlight topic areas that should be assessed in order to assure reliable scientific evidence from the anthropological field as well as encourage communication among professionals (Nakhaeizadeh et al., 2014; Christensen & Crowder, 2009).
In conclusion, the outcome of the legal reform initiated by Frye and continued under Daubert has permanently changed the way scientific justice is practised. ‘Junk sciences’ are now being held responsible for a lot of miscarriages of justice and unreliable scientific evidence given in Court. Forensic anthropology is also considered as a possible candidate on this matter as its methodology is sometimes considered too subjective, affected by contextual cognitive bias and mainly relying on methods based on non-specific populations. There is an obvious need for additional studies to be done in order for forensic anthropology to eradicate ‘junk’ scientific opinion. Fortunately, there are some pertinent changes in the field that are being brought forward in hope of adjusting the way forensic anthropology is practised. Nevertheless, forensic sciences might not be the only problem in the current juridical situation. In fact, various scholars studying the issue of ‘junk science’ have wondered if having a jury was actually part of the problem. Obviously, there is a need for a ‘better’ forensic practice whether there is a jury or not. However, because forensic evidence in other Commonwealth countries is presented and heard by the Judge rather than jurors, an unreliable opinion is simply discarded by the Court and no decision is made based on improper scientific knowledge (Bernstein, 1996). On that note, a change in the way scientific evidence is heard might need to be studied.
...(download the rest of the essay above)