How did the body become a source of objective knowledge for health sciences?

In this essay, I argue that the body becomes a source of objective knowledge through its visual representation as an object. First, I define the body as understood by the health sciences. Second, I trace the history of representations of the body in biomedicine, and locate the transitions of visual representations of the body in the medical education of physicians from dead to live, artistic to mechanical reproduction, contemporising this discussion in the 21st century through technology, particularly radiology. Next, I argue that representations of the body, still need to be perceived, and hence the body as objective knowledge is an illusion. Lastly, I situate this discussion against the backdrop of subjective ways of knowing the body.

The Body, Scientific Objectivity and the Health Sciences

Scientific objectivity is ‘blind sight, seeing without inference, interpretation or intelligence’, and ‘aspire(s) to knowledge that bears no trace of the knower’, where ‘knowledge is unmarked by prejudice or skill, fantasy or judgement’(Daston and Galison 2007:17). The human body as understood by the biomedical health sciences is a corporeal entity, with ‘the real(i.e. visible, palpable) from unreal’(Scheper-Hughes and Lock 1987:6), and that can be solely understood through the mere mechanistic entity, organisation of system and biological processes of anatomy, histology, physiology, pathology(Good 1993:65). To the naked eye, gross anatomy is a ‘classification or taxonomy which is itself a model; learning anatomy means grasping a three-dimensional structure and naming its parts’(McDonald 2013:129). The body as an object is a technical problem to be solved through science, that inherently does not carry metaphoric meaning or emotional ties(Lock 2018:228), and is marked against standardised norms(Lock 2018: 38). Hence, in the body becoming an source of objective knowledge, the body too, bears no ability of judgement, the mind is separated from the body as in cartesian dualism(Scheper-Hughes and Lock 1987:6).

In asking how did the body become a source of objective knowledge, is also to ask how did the health sciences understand the body in the way it does. Byron Good, American anthropologist, utilises ethnography of the education process of Harvard medical students learning medicine, as a means of understanding ‘how medicine constructs the ‘objects’ to which clinicians attends’(1993:65). In particular, the anatomy lab is a ‘ritual space’(Good 1993:72). The body surfaces may convey personhood, but as the skin is removed, the interiority of a person is understood as internal organs, such as the liver, or kidneys, as opposed to their thoughts, experiences, personality. Aspects of the transformation of a person to cadaver are ‘deliberately occluded’, where there are discrete entrances of buildings where bodies can be brought in, with the preservation process effected by technicians in the backrooms hidden(McDonald 2013:138). In this process of learning anatomy from dead bodies, educators tell students to suppress their strong emotional responses and to be ‘Be objective’ and ‘Distance yourself’(McDonald 2013:139). In this process in becoming physicians, medical students themselves too undergo a separation of mind and body.

Representations of the Body in Anatomy Atlases

Representation, as described by Michael Taussig in Mimesis and Alterity, starts first with mimesis, a two-layered process of reproduction, and then subconscious perception of the object by the perceiver(Taussig 1993:22). The first part of the process requires a ‘mimetic faculty’(Benjamin 1933) or the ability to create a model of an object(Taussig 1993:18). Mimetic faculty is made possible with the development of mimetic machinery which allows for the recording and distribution of media(Taussig 1993: 20). The second part of the process requires a perception by the perceiver, where contact is made with the body, through the process of seeing, where a copy is registered onto the retina(Marx 2004 quoted in Taussig 1993:22). Crucially, the model ‘gains the power of the original, the representation the power of the represented(Taussig 1993:xv) and the capacity to Other(Taussig 1993: 19). While Taussig does not directly define power, the implication of the power he describes is closest to Max Weber definition of power as ‘the probability that one actor within a social relationship will be in a position to carry out his own will despite resistance, regardless of the basis on which this probability rests’(1922)1968:53). Power is gained as the subconscious perception opens up the optical unconscious, where the perceiver ‘comprehends hidden details of familiar objects’(Taussig 1993:24), then building habits of knowing, switching on the automatic pilot of ways of relating to the object.

In Bodies and Cadavers, Maryon McDonald traces the ways bodies are turned into an anatomical state (McDonald 2013:131), first with simplification ‘purification and filtration’, and then standardisation. These diagrams are a model of the body as an object, and is meant to be a definitive representations of the body. Accordingly, the consolidation of diagrams in anatomy atlases are viewed as an important ‘three-dimensional topographical map’(McDonald 2013:134). Atlases are used to train medical students on how to perceive, and see(Good 1993:71) the body. In the anatomy lab, diagrams are made from dissections, and these diagrams are then referred to during dissections to identify structures(McDonald 2013:134) by medical students to then learn how to identify normal or deviant bodies.(Figure 1) Atlases hold great power in creating habits of knowing, ‘not only do images make the atlas; atlas images make the science’, and are ‘dictionaries of the sciences to the eye’(Daston and Galison 2007:22).

Radiology as Mechanistic Objectivity

Radiology started in 1895 when X-rays were discovered. In just a century, radiology developed rapidly as a discipline, and is now central to the practice of 21st century clinical medicine(Thomas and Banerjee 2013, 193-204). In the 1920s, training was required for personnel to report on images. In the 1980s, cross-sectional imaging, such as computed tomography or magnetic resource imaging, enabled the viewing of internal organs, rather than just bones, and became ubiquitous with the shift from analogue technique into digital storage. Since the 1990s, cross-sectional imaging emerged as the mainstay of diagnosing acute conditions.

Radiological images differ from representations of the body in Anatomy Atlases developed from cadavers, in that radiological images required the development of radiological technology. In anatomy atlases, mimesis of the body occurred through the mimetic faculty of a human artist’s rendition of the cadavers. The mimetic machinery, as with in the 20th century, involved cameras to photograph dissected cadavers, and the printing press in order to print and distribute atlases. In contrast, mimesis of the body by radiology created further distance between the perceiver and the body, as reproduction is done by the scan machine, rather than a human artist.

Hence, radiology as a mechanical reproduction, as defined as the use of instruments and technology(Daston and Galison 2007:120), was a form of mechanical objectivity, bringing the body closer to scientific objectivity as ‘blind sight’(Daston and Galison 2007:17). In radiology, the mimetic faculty eliminated the human, who might have the ‘temptations of aesthetiscs to … schematise, beautify, simplify’(Daston and Galison 2007:120). Instead, the scan machines are ‘repetitive, mindless, automatic’ and achieved ‘mechanical objectivity’ as the ‘positive ideal of the observer’, where it was ‘patient, indefatigable, ever alert, probing beyond the limits of the human senses’, counteracted from ‘all-too-human weaknesses of… inattention, idleness or dishonesty’(Daston and Galison 2007:137-9).

Radiology as Unfaithful Reproduction, The Inside Alive

Positron Emission Tomography(PET) scans are one of the latest development in cross-sectional imaging, which involves a CT scan but where patients are injected with radioactive nuclides in order to capture molecular activity cells(NIH, 2022). PET scans can be used in cancer staging, diagnosis of occult infections and inflammations. In Brain Scans and Biomedical Identity(2004), Joseph Dumit, American anthropologist, analyses how the PET scan have come to be representation of human bodies through interviews with world leading experts in radiology and PET who come from an interdisciplinary background from nuclear medicine, medicine, physics. Dumit also traces the history where PET scans have also been used to depict the interior of a human, as identity and personality. PET scans differ from cross-sectional imaging of CT scans or MRI because of the use of colour(Figure 2), and its representation of not just the insides of the internal organs of the corporeal body, but also that of molecular processes or life itself.

PET scans are intentionally an unfaithful reproduction of the body. The process of reproduction, or the creation of a model is not just a ‘faithful likeness’(Taussig 1993: 62), as the process of reproduction will create alterity, or introduction of differences between the object and its model. Anatomy atlases were created with the aim to be a ‘faithful representation’ as close as reality to the physical body(McDonald 2013:134), depictions of the body were coloured are similar to that of the structures that they were presenting, for instance, arteries as red, veins as blue, or muscles as flesh coloured(Standring, Anand and Tunstall, 2021). In contrast, the reporting of PET scan images require the presentation of colour through ‘pseudo-colour’.

Pseudocolour, is used to report the activity of the radioactive nuclides, which itself is a representation biochemical processes within the body. Pseudocolour ‘exaggerates and may distort the information that is in the imaged data’(Dumit 2004:92). The presence of colour is further development of the grayscale images in CT and MRI scans. The debates over colour schemes in reporting PET scans ‘concern clarity versus a notion of fidelity’(Dumit 2004:93). The choice of pseudocolour can ‘create a significant visual shift, rendering a small change in numerical value as a solid boundary between what now appears to be distinct regions’, and ‘the arbitrariness of the colors reinforce the sense that these regions are internally coherent’(Dumit 2004:93). As Taussig argues, the power of that which the representation reflects is not based on its faithful likeness, but the power as ‘captured only by means of an image, and better still by entering into the image. The image is more powerful than what it is an image of(Taussig 1993:62).

Phelps: You know, another way to approach the explanation is to forget about PET initially and focus on the problem: That is to be able to take a camera and just watch. Inside the body is all this biology that we know is going on. You take food in, you eat it, and it becomes nutrients for your cells.(Michael Phelps as quoted by Dumit 2004:2)

PET scans were a significant development from previous radiological techniques, because they no longer merely represent the body as anatomical structure, but also represent molecular processes. Cross-sectional imaging already provided the possibility of viewing the interior of the body while the body is still alive. PET scans push that further by seeing life itself, by viewing the molecular processes that lead to life. This significantly differs from previous ‘medical education still begins with the dissection of a cadaver, just as the clinical case ends in the pathologist’s lab, and in between, the living patient is often treated in a cadaverous or machine-like fashion’(Leder 1998). The dead body, Drew Leder, a philosopher and medical doctor, points out the contradiction in that physicians learn to treat live bodies through dead bodies, reinforcing the cartesian dualism in medical practice, where the human is a body as a corporeal entity. With PET scans, even though the body is depicted as life, the living body is now presented as an object.

Trained Judgement and the Illusion of Objectivity

What’s the difference between the 40 images? Which is normal, which has a tumor, and which has indications of stroke? Actually they’re all the same image of a healthy normal volunteer— just displayed with different color scales.(Brian Murphy as quoted by Dumit 2044:94)

With mechanical objectivity, the burden of representing the body lay in the radiological image itself. However, in the early 20th century, physicians who used radiological images noted that the multitude of variants, both normal and pathological, cannot be held in any single representation. The purity of blind sight for scientific objectivity, where the human was eliminated, came at the cost of acknowledging the essential role of the readers’ response: the human capacity to render judgement. Trained judgement became a way to supplement mechanical objectivity to make hypotheses based on the images, for research and explanation, even images from the most sophisticated instruments (Daston and Galison 2007:312). Trained judgement, is defined by the ability to ‘synthesise, highlight and grasp relationships in ways that were not reducible to mechanical procedures’(Daston and Galison 2007:312).

A century after the initial x-ray images, these discussions are still being had with PET scans, despite significantly more sophisticated technology. The differentiation of normal and pathological on the scans can be arbitrary based on the grading of pseudocolour scale (Figure 3). Without any standardisation of pseudocolour across institutions, or nations, the slight alteration of pseudocolour scale can completely change a diagnosis, indicate a pathology where there is none. This inability to provide standardised scales have meant that the trained judgement is not just in interpreting the images, but also choosing the scales which would influence the interpretation. “People have a tendency, of course, to use the scales that emphasize what they like to emphasize”(2004:94) Dr Michel Ter-Pogossian, professor of radiology at the Washington University School of Medicine, who is also known as the Father of the PET states “Parenthetically, the pictures that are particularly attractive that you have seen in general are fairly heavily doctored, in the sense of making them more attractive than they should be”. Hence, contrary to the expectation of mechanical objectivity, PET scan images can in fact be used as to favour a certain judgement, under the guise of objectivity.

Conclusion

In this essay, I have argued that the rise of imaging technology to depict the body as object was responding to the demand for scientific objectivity, however, the impossibility of eliminating the perceiver who sees, with its human flaws, means that pure objectivity might never be achieved. Yet, I do not advocate for the reversal of radiological technology, but instead the recognition that ‘subjective’ forms of knowing from the ‘lived body’ are valuable too. Technologies, including radiology, are being viewed as replacing the senses, and the physical examination is seen as being at threat of being lost(Harris 2016:33). The physical examination where medical students learn consisting of inspection, palpation, percussion and auscultation,(Good 1993) are a form of bodily technique(Mauss 1973) in the diagnostic process too. ‘The lived body is not an abstract entity… It's the body as it's felt, experienced, and lived through the richness of daily existence’ Leder(1992). Moving into a definition of the body, which is more than just a corporeal entity, with the sensations, emotions, and perceptions intertwined, reminds physicians that objective ways of knowing the body are incomplete. With the ubiquity of radiological imaging, doctors can gain knowledge of the body, without touching the physical body at all. Hence, rather than radiological imaging as a representation of the body, radiology, as a mimesis of the body may have instead, merely, as Taussig states, taken power from the represented, even if that is from our very flesh.

Figures

Figure 1

(above) A labelled photograph of a dissected head and neck in the sagittal view used in the training of surgeons, from Overstall and Zaveri, Get Through MRCS: Anatomy 2E.  Page 37.

(below) An artist rendition of the head and neck from a key textbook used in medical education of medical students. Drake, Vogl, and Mitchell, Gray’s Anatomy for Students E-Book. Page 1050.

Figure 2

PET scans can be combined with existing cross-sectional imaging such as CT and MRI. A and B shows CT and MRI scan modalities when used individually. C and D show CT and MRI scan modalities when overlaid with PET modality. Appenzeller, P., Mader, C., Huellner, M.W. et al. PET/CT versus body coil PET/MRI: how low can you go?. Insights Imaging 4, 481–490 (2013). https://doi.org/10.1007/s13244-013-0247-7

Figure 3

Images from a PET scan of a patient with no pathology with different scales for pseudo-colours. Colour plate 12 from Dumit, Picturing Personhood.

Bibliography

Benjamin, W. (1933) ‘On the Mimetic Faculty’, in M.M. Lock and J. Farquhar (eds) Beyond the Body Proper: Reading the Anthropology of Material Life. Duke University Press, pp. 130–132.

Daston, L. and Galison, P. (2007) Objectivity. Zone Books. Available at: https://doi.org/10.2307/j.ctv1c9hq4d.

Good, B.J. (1993) ‘How medicine constructs its objects’, in. Cambridge University Press, pp. 65–87. Available at: https://doi.org/10.1017/CBO9780511811029.005.

Harris, A. (2016) ‘Listening-touch, Affect and the Crafting of Medical Bodies through Percussion’, Body & Society, 22(1), pp. 31–61. Available at: https://doi.org/10.1177/1357034X15604031.

Leder, D. (1998) ‘Chapter 4: A tale of two bodies: the Cartesian Corpse and the lived bod’, in D. Welton (ed.) Body and flesh: a philosophical reader. Malden, Mass: Blackwell publishing, pp. 17–35.

Lock, M.M. (2018) An anthropology of biomedicine. Second edition. Hoboken, New Jersey: Wiley Blackwell.

Marx, K. (2004) Capital: Volume I. Penguin UK.

Mauss, M. (1973) ‘Techniques of the body’, Economy and Society, 2(1), pp. 70–88. Available at: https://doi.org/10.1080/03085147300000003.

McDonald, M. (2013) ‘Bodies and cadavers Maryon McDonald’, in G. Evans et al. (eds) Objects and Materials. Routledge, pp. 146–161. Available at: https://doi.org/10.4324/9780203093610-19.

NIH, N.I. of H. (2022) Positron Emission Tomography (PET). Available at: https://clinicalcenter.nih.gov/pet/about_pet.html (Accessed: 31 May 2024).

Scheper-Hughes, N. and Lock, M.M. (1987) ‘The Mindful Body: A Prolegomenon to Future Work in Medical Anthropology’, Medical Anthropology Quarterly, 1(1), pp. 6–41. Available at: https://doi.org/10.1525/maq.1987.1.1.02a00020.

Standring, S., Anand, N. and Tunstall, R. (2021) Gray’s anatomy: the anatomical basis of clinical practice. Forty-second edition. New York: Elsevier.

Taussig, M.T. (1993) Mimesis and alterity: a particular history of the senses. New York ; Routlege.

Thomas, A.M.K. and Banerjee, A.K. (2013) ‘Review and the future’, in A.M.K. Thomas and A.K. Banerjee (eds) The History of Radiology. Oxford University Press, p. 0.

Weber, M. (1968) Economy and society: an outline of interpretive sociology. New York: Bedminster Press.