Preliminary Draft
For the use of the participants in the workshop "Writing Science" only.
Please do not cite or distribute.
Copyright 1997 by John Bender.

Click Thumbnails for Larger Images

John Bender

"Opening Up a Few Corpses, 1795-1995"

This talk presents two propositions about the history of virtual representation of the human body in medical science since the Enlightenment. By "virtuality," I mean something like "the staging of presence through illusion or hyper-illusion." Today, I'm chiefly concerned to illustrate the first of these propositions, which is that the history of bodily virtuality is a history of media forms with both continuities and discontinuities. I'll make this point by presenting a rather stark contrast between eighteenth-century anatomical illustration and some new types of body imaging used in medicine today. My second proposition will be that--with regard to debates about the "post-modern" episteme as it relates to the Enlightenment--the domain of anatomical representation reveals, in its earlier forms, aspects of the "virtuality" or simulation we associate with so-called post-modernity itself. The obvious question, then, though one I can't pursue it in detail, is whether the real break into a post-Enlightenment episteme in fact has yet occurred since we can see a simultaneous preservation of early forms within the paradigm-breaking innovations happening around us today.

Let's start by focusing on some engravings in the realist mode typical of anatomical atlases in Europe during the scientific revolution, but especially those from the later seventeenth century and eighteenth centuries. There were stylized departures from this realist mode, such as the classicizing manner of Albinus:

But two earlier 18th-century plates from Godefridus Bidloo's atlas of 1685 and a later one from William Hunter's Atlas of the Gravid Uterusof 1774 can serve here to sum up this the realist anatomical "medium."

In fact, Hunter himself associated the mode of his own atlas with Bidloo's and noted that they stove for a "close representation of nature. . . finished from a view of one subject" and, were therefore, "often . . . somewhat indistinct or defective in some parts." Works like those of Albinus and the Eustachius, whom he specifically names, he called figures of "fancy." Hunter declared that "the one shews the object, or gives perception[my italics], the other only describes or gives an idea of it. A very essential advantage of the first is, that as it represents what was actually seen, it carries the mark of truth, and becomes almost as infallible as the object itself." Since such plates attempt to present crystalized, clarified, and certifiably accurate versions of what one might actually see in person in a laboratory, I will style this as the medium of the "modest witness." This kind of witness is observant only--not directly interactive with the object observed. These images evoke "personal" subjective presence in the experience of the viewer through the combination of virtuosic aestheticism with a certain matter-of-fact brutality evoked by the inclusion of the tools of experimentation.

Note here the crucial idea that the evocation of bodily presence in the image ultimately works to objectify the subjective presence of the observer by, paradoxially, placing him (and usually in this period the scientific observer is male) in the impersonal position of the modest witness who is capable of noting clamly even the shocking spectacle of bodily dismemberment. The projection of publicness is a crucial part of the "truth" and "realism" of this imagery, as Hunter attests when he declares, in the Preface to this Atlas of the Gravid Uterus,that "every part" of his dissections "was examined in the msot public manner, and the truth was thereaby well authenticated."

Next let's move to the 18th-century medium that I would call "virtual witnessing." Here, I'm purposely ringing a change on Shapin and Schaffer's terms in Leviathan and the Air-Pumpby teasing apart the idea of the "modest witness" from that of "virtual witnessing." By the latter I mean representations that a achieve a hyper-reality by showing more than can ever be seen by an actual witness. As instances I'll offer two models in wax from the Fontana school of the last part of the eighteenth century.

In the examples shown here, in particular, lymphatic structures that are impossible to dissect in their totality in a single specimen appear fully displayed and anatomic structures are positioned in ways that defy gravitation.

Other, more famous, models of the Fontana school come apart in sequence to mimic actual dissection and thus offer a version of the interactive, three-dimensional virtuality that we now associate so strongly with post-modernity.

In earlier media formations, similar effects were attempted with layered, fold-out engravings a little like children's pop-ups.

Indeed, certain early anatomical engravings incorporate and regulate the observer position by assigning it to the object of the dissection himself.

Note that a crucial difference between the Fontana models and present-day virtuality is that the interactive dimension is only one-way in the wax models: a student may disassemble them but the model cannot change their shape in reaction to the movements of the student, nor do they simulate the texture and resilience of actual tissue. But, even now, as especially for their period, these models still edge towards the kinesthetic dimensions of simlulation through which motion photography and computer-based virtuality so dramatically stage presence.

The kinesthetic aspect is crucial, I think, in imagining these earlier anatomical images as exhibits in the history of "virtuality." In the case of the models, the sense of touch is vividly evoked even when one in fact does not touch the model: this is sense of "touchability" crucial to the what I am calling kinesthesia in earlier imagery, just as it is crucial to modern virtual reality systems. It is present in attenuated form, I believe, even in the Bidloo and Hunter engravings, for, the "public" witness whom Hunter's stresses (what I am calling the "modest" witness) is a potential toucher.

The surviving eighteenth-century materials that to my mind produce the most astonishing effect of kinesthetic presence are those produced by Honoré Fragonard at the Ecole Vétérinaire in Paris. Photographs do not make clear the crucial truth--at once evident when one sees these figures in person--that, though wax is the medium of preservation and illucidation, these figures, unlike the Fontana models, are based upon actual human and animal tissue.

Remarkable is the use of glass eyes to make the standing figure come "vitually" to life. Glass eyes are common in wax figures but seem all the more powerful here because these figures, if they lived, would assume aspects like the Frankenstein monster. Glass eyes are an important part of the story, though not one I fully understand at this point. Why are they needed here in the Fragonard model. More puzzling still is why they are needed in this head of a child from William Hunter's collection of anatomical specimens now in Glasgow:

I would venture to suggest that they are a kind of anticipation of motion photography and the coming to life so crucial to it.

In advanced modern virtual reality systems, one gets, by contrast with eighteenth-century "virtuality" and with conventional motion photography, full two-way interactivity--in real-time--with haptic feedback. I can't show you this since it would take a million dollars worth of equipment but I can project a virtual fly-through of the colon that is reconstructed from the 1800 photographic slices, plus CT and MRI images, slices of an actual man that make up the data set of the "Virtual Human." This data set is accessible on the web-site of the National Medical Library in Bethesda, Maryland under the title of the "Visible Human." Note that this animation is actually a range of data sets oriented to an observer-centered staging of presence. These data do not form this image in their unprocessed form.

Animated Colon (QuickTime Movie)

A fully interactive version would allow us, or the surgeon in training, to stop, to "feel" the imaged tissue as if it were "real," and even to excise imagined/imaged growths. Oddly close to this fly-through was a terrifying machine placed on exhibit in London for five shillings per view during the year 1733 by the surgeon Abraham Chovet. In order to give a faint idea of it, I show a slide from the Bidloo atlas:

It was a new "figure of anatomy" that made "the Circulation of the Blood . . . visible, through Glass Veins and Arteries, with the Actions of the Heart and Lungs; as also, the course of the Blood from the Mother to the Child, and from the Child to the Mother."

The figure represented a Woman gone eight Months with Child, chained down upon a Table, supposed to be open'd alive of which the two principal Cavities are laid open, viz. the Breast and the lower Belly, which are divided from each other by the Diaphragm or Midriff. In the Breast, the Heart which is placed to move for carrying on the Circulation is seen between the two Lobes of the Lungs, which likewise move as in Breathing; from the Heart are seen going out the two principal Arteries of the Body made of Glass; the one leading to the Lungs, and the other towards every Part of the Body, viz.. toward the Head and Arms, whilst the Trunk is continued on the Back Bones through the Breask, and passing the Diaphragm into the lower Belly, divides itself to all its Parts, and then goes to the Legs. The same Number of Veins made of Glass accompany the Arteries, join in one Trunk, and open into the Heart again: Through the Arteries, a red Liquor, in imitation of Blood, is seen to move from the Heart to every part of the Body; and by the Veins returning from every Part of the Body to the Heart again. In the lower Belly, in a Cavity called the Pelvis or Bason, the Child is seenb in the Womb, with the After-birth and Navel-string made of Glass, consisting of two Arteries and one Vein arising from the After-birth by many Branches, which terminate in one a considerable Length, and enter the Navel of the Child, through which the blood is seen to pass; from the Mother for its Nourishment. The two Arteries are likewise seen to rise out of the Navel of the Child, twisted round the Vein in its Course towards the After-birth, where it divides itself into a great Number of Branches, in the same Manner as the Vein: Thro' these Arteries the Blood is likewise seen to return from the Child to the Mother after Nourishment.

Chovet was fully aware of the dangerous representational terrain--I would say the frightening yet hypnotic display of presence that he had produced--for he placed an "Animadversion" at the start of his book saying that, because his model was produced for instruction about the circulation of blood "is was absolutely necessary that it should represent a Woman, supposed to be opened when alive, because thee are all vital Functions, which are not exerecised in a body when dead. Therefore it is to be hoped that nobody will make objection to the Representation, which would carry with it an Idea of the highest Barbarity and Cruelty, had it ever been put in Practice upon any Humane Body." What I'm suggesting here is that while this model obviously was different in every technical respect from the colon fly through based on the Virtual Human, it was aspiring to kinds of data display that were analogous, including kinesthetic effects produced in real-time dimension. Actually, there are important overlaps between the old anatomical atlases and the Visible Human, though they do not produce visual effects quite a spectacular as the colon flythrough. Traditional atlases often presented coronal sections of the body (what we would call "slices" in non-technical English). One of the primary datasets of the Visible Human are the 1800 photographs made as each slice was taken through the cryogenically processed body (the new female Visible Human has 5000 slices). MRI and CT data, of course, are rendered in this same form--perhaps under the influence of traditional atlas forms, though they could be processesd into other kinds of display. The difference is that the slices of the Visible Human are infinitely finer and more numerous than those in any traditional atlas and, therefore, can be assembled into another kind of flythrough, which I now will demonstrate:

Virtual Skeleton (QuickTime Movie)

A final kind of imagining that runs in tandem with computer based virtuality is what I would call "intra-structural" virtuality--following Paul Virilio in his Art of the Motor.A primitive form of "intra-structural" virtuality can be seen in the surgery tape I'm about to show in which gall-bladder surgery is conducted by a physician who never sees the inside of the body directly but only through magnified video imaging. The haptic element here is direct since the instruments are manipulated by the surgeon's hands.

With regard to my second proposition, then, one might say that a true epistemic break really will have begun when computer-based virtuality comes together with "intra-structural" imaging. This would occur in surgery, for instance, when the doctor's interaction with a three-dimensional virtual image--with haptic feedback--included: 1) interaction with a data-set that superimposed models and armatures, which one might liken to the classical anatomical atlas, upon the peculiar anatomy and real-time reactions of the patient's tissue in the surgical situation; and (2) when the surgeon's interaction with the virtual image guided instruments that the surgeon's hands never touched and which, on Virilio's model, would be minute "nano" instruments located within the body and run by tiny motors. Data sets relevant to the observer would then have become intermediary to the illusion of presence for the operator but input from the operator based on this illusion might be only one factor guiding the surgical instruments. For instance, atlas-like templates derived from standardized data based on many bodies might be combined by computer with the CT and MRI data taken from the patient under surgery, to limit or even override the choices made by the human surgeon in his presence-oriented version of reality. The encompassing virtuality of the surgery from the human surgeon's point of view would be only one dimension of the larger project. Brain surgery by robotic lasers already is headed in the direction I am envisioning--a direction in which the actions of a human surgeon become one of several data sets processed to excute the surgery.

When this occurs, we might in fact have the sharp epistemic break that Linda Williams, in Hard Core,prematurely attributes to the invention of motion photography. She certainly is right that this invention intensified the field of the visible and, with it, the capacity to stage presence through the simulation of pleasure. Motion photography of the body, like the 18th-century innovations in anatomical representation that I have sketched here, preserves such "presence"-generating elements as the (1) "gaze" of single-point perspective, which idealizes the actual observer as a modest or virtual witness; (2) the sense of literal touch as opposed to "touch" mediated through a data set. In these respects, the earlier examples I've offered--especially Chovet's "living" model--reach toward the effects more fully achieved in motion photography.

What is a stake here, it seems to me, are questions about the history of "presence." The segment of this history that we call the "Enlightenment" is one in which the older "magical" forms of presence that inhabited the body with "spirit"--obviously I'm thinking here of the eucarist and saintly or demonic possession, as well as the forms of absolute kingship carried in the idea of the king's two bodies--shifted into forms of presence that inhabited the body with the correlative opposites of disembodied "objectivity" and hyper-embodied "pleasure." The various robotic forms that have become, or will shortly become, feasible in medicine and other domains would seem to drain way presence of the Enlightenment kind. The surgeon interacting with the joint data-sets that form the virtual image through the manipulation of which she guides the nano-instruments, could herself become a data-set or robot--as already is happening in brain surgery.