A clinical example (for MJ)

I'll quote a case fragment that was reported by a friend a while ago. Back then, endoscopy wasn't widely available, and if you wanted to get one for a patient, you'd have to get a seriography first. That's a variety of x-ray exam where the patient ingests the contrast medium and a series of plates is made as the contrast slips through the digestive system, leaving an imprint of the inner lining of the digestive tube on its wake. If the x-ray image suggested an ulcer, than the patient would be sent to endoscopy.
Due to some clerical error, one such patient ended up at the endocrinology department, instead. The endocrinologist examined the patient, and came out with a typical story of hyperthyroidism (sweats, tremors, anxiety), and asked a number of tests (which came all negative). When the case was later discussed in a clinical session, a more experienced radiologist stated that there was no abnormality in the original x-ray plates, it was just an artifact.
Those were all trained physicians in a university hospital, this is not an example of bad medicine or medical error; the moral of the story, here, is that the training of the endocrinologist directed his perception towards a set of symptoms, which fit the patterns he'd come to expect in his practice; before him, the general clinician was also directed to diagnose a disease (i.e., some organic disfunction secondary to a biological lesion), and did not pay attention to the subjective aspects brought by the patient (this is a point that I address with more details in a recent paper, the reference, if you'd like to take a look, is CAMARGO JR, K. R. ; COELI, Claudia Medina . Theory in Practice: Why Good Medicine and Scientific Medicine are not Necessarily the Same Thing. Advances In Health Sciences Education, v. 11, n. 1, p. 77-89, 2006).
The point here is that, instead of a disease that "was always there" and is passively perceived as such by the doctors, what we have is an active process of categorizing ailments that in this sense creates the disease.
Note that in saying this I'm not claiming that there is anything wrong with that.
I'll get back to this in my next post on classification.
But keep the questions coming!

And by the way.../E por falar nisso...

Happy/Feliz /Joyeux

2007

Recommendation/Recomendação

The links below/os links abaixo:

http://scienceblogs.com/effectmeasure/2006/12/the_chicken_soup_challenge.php
http://scienceblogs.com/effectmeasure/2006/12/wherein_i_respond_to_oracs_cou.php

EffectMeasure is one of the blogs that I read regularly. The two posts above are related to the issues I'd like to tackle here, and I wholeheartedly recommend those in particular.

EffectMeasure é um dos blogs que eu leio regularmente. Os dois textos acima estão relacionados aos asuntos que eu gostaria de abordar aqui, e eu os recomendo fortemente, em particular.

Are you still there?/Vocês ainda estão aí?

(em português mais abaixo)

Just to stir things up before I come back with a real post. I have to admit to a bit of disappointment with the response (or lack of) to this initiative. Probably I had some unreal expectations to start with. :)
Nevertheless, I'll keep on trying to prod you people. I don't know to what extent that plays a role in the lack of comments, but I can see that some of the language employed here can look a bit impenetrable at times. As I stated on the opening post of this blog, much of what I'm writing here is a shorthand for stuff that I intend to develop further, so if it looks hurried, half-baked, not well expounded, well, it is indeed hurried, half-baked and not well expounded. Sorry! But how can we improve on that? Don't hesitate in saying "I do not understand", or "this is not clear". As someone involved in teaching for almost thirty years (yes, I started early), all I can say is that the worst kind of question is the one not being asked... the better way to ensure we can create some degree of understanding is to point out what is not clear, and take that as a starting point (and Ben, no, "everything" is not an option ;) ). So, I'm game, who's gonna take a shot?

And an afterthought: writing in both languages is a pain in the *. I would like some feedback on that as well, please.

-------------------------------------------------------------------

Só para agitar um pouco as coisas antes de eu retornar com um texto de verdade. Eu tenho que admitir um certo desapontamento com a resposta (ou falta de) a essa iniciativa. Provavelmente eu comecei com expectativas irreais. :)
Ainda assim, vou continuar tentando cutucar vocês. Eu não sei em que grau isso pode ter influenciado na falta de comentários, mas eu posso ver que a linguagem utilizada aqui pode parecer meio impenetrável às vezes. Como eu escrevi no texto de abertura desse blog, muito do que eu pretendo escrever aqui é um rascunho abreviado de coisas que eu pretendo desenvolver mais, portanto se parece apressado, mal-feito e não muito bem explicado, bom, é realmente apressado, mal-feito e não muito bem explicado. Perdão! Mas como podemos melhorar isso? Não hesitem em dizer " não entendi" ou "isso não está claro". Como alguém envolvido com ensino há quase trinta anos (é, eu comecei cedo), tudo o que eu posso dizer é que o pior tipo de pergunta é a que não é feita... a melhor forma de garantir que nós possamos criar algum grau de entendimento é mostrar o que não está claro, e tomar isso como ponto de partida (e Ben, não, "tudo" não é uma opção ;) ). Então, estou pronto, quem quer mandar alguma coisa?

E mais uma coisa: escrever em ambas línguas é um pé no *. Eu gostaria de algum retorno sobre isso também, por favor.

Giving another shot/Tentando de novo

When I started this experiment I said I would use only the English language in my posts. Given the fact that this blog so far has received only one comment, I will try something different, even though I am not sure whether the reason for the low turnout is actually related to the language of the posts. So, from now on the posts will be bilingual - let us see how this works out.

Quando comecei este experimento eu disse que utilizaria apenas o inglês nos meus textos. Tendo em vista que este blog até o momento recebeu apenas um comentário, vou tentar algo diferente, mesmo não tendo certeza de que a razão para a escassez de respostas seja de fato a linguagem dos textos. Sendo assim, daqui em diante os meus textos serão bilíngües - vamos ver como isso vai funcionar.

A small experiment

Given the interdisciplinary nature of both staff and student body where I teach, my graduate classes usually have a mixed attendance, some people having been trained as health care professionals and others not.
For a number of semesters I ran a small "experiment" with the students enrolled in my seminar, which consists of showing them a scribbling more or less like this:

and then asking what they think it is. Almost invariably, the health trained people said "it's an EKG", while the others gave different answers (more frequently, "it's a squiggle").

This, in my view, illustrates one of the most important of Fleck's (reader warning: get used to this name!) ideas: the thought style, which he defined as "a definite constraint on thought, and even more; it is the entirety of intellectual preparedness or readiness for one particular way of seeing and acting and no other". In this case, the health care people, having previously dealt with electrocardiograms, interpreted my doodle (which, incidentally, was made as casually as possible, with no regards to realism) as one of them, whereas the others, lacking that training, didn't make the same association. In fact, we could go a step further backwards; actually what I had shown them was a line of ink on a paper, but due to shared assumptions about what I was asking them to consider, that is, whatever I was representing with that line, they ignored the material support of the image.

Now, to the point. To the trained "eye", we do not have a doodle representing an abstracted, idealized version of the inscriptions produced by a machine that, we are told, registers the electrical activity of the heart, but the EKG, a thing in itself. Any thought of a clear-cut boundary between "immediate perception" and its interpretation is severely compromised here.

Before I proceed, I would like to invoke another example. Consider now the sorts of optical illusions that can be interpreted in two ways, like this image:




This can be seen either as a saxophone player or a woman's face (can you see both?). These are also the basis of another kind of experiment, but with my undergrad (medical) students. We usually show them a bunch of those images in a class activity intended to work as a device to provoke some discussion about perception and criticism. The relevant thing here is that more often than not most students cannot see the two images, and need to be taught to do so; conversely, and also very important, once they learn how to see the two images, there is no way to recover the original, untrained, perception of the drawing.

These observations, paired with Fleck's concepts, have in my opinion important implications for medical practice and its study. I will just illustrate two. Medical reasoning revolves around recognizing certain patterns and categorizing them as one of a number of already catalogued diseases; because of that, patients discourses will almost inevitably be filtered to fit the nearest disease pattern, meaning that, for instance, existential considerations that might need attention could go unnoticed.
The other aspect to consider is the huge array of tests and exams deployed in modern medical diagnosis, including the already mentioned electrocardiograms. Again, through training, they gradually become "things" in themselves, and the mediating processes that take place between whatever bodily phenomenon is being tracked and the image or number the method produces is lost from sight; not only the exam or test becomes a black box, as Latour puts it, but its results are equated with the very "thing" they are supposed to portrait.
Yet again, another reason for taking sharp distinctions between "things" and "ideas" with a grain of salt.

Laboratories.

One of the things I intend to use this venue for is to put down in writing some loose ideas still in a state of flux (hence the name of the blog) before they go from loose to lost. :) So expect some half-baked arguments, free association and the like; the full-fledged thing, if it ever arises, will be a paper or book chapter, this is just a springboard. But your comments are still very welcome!
In this post I will jot down a few general ideas about laboratories and their role in producing scientific knowledge and artifacts. There are four texts that I consider extremely relevant in this regard (full references at the end of the post): the second section of the fourth chapter of Fleck's book (this is an author that is going to figure a lot in this blog); Gooding's "Putting agency back into experiment"; Hacking's "The self-vindication of the laboratory sciences" and finally Latour's "Give me a laboratory and I will raise the world".
Coming from fairly different angles, these four texts share some common traits and complement each other. Fleck discusses, based first on one of Wasserman's experiments and then on one of his own, how a certain predisposition to perceive things in a certain way (what he defined as a "thought style"), a set of practices and collective interactions shape up the laboratory "facts", from fuzzy, incertain starting experiments. Gooding elicits the minutiae of decisions and adjustments that make up the life cycle of an experiment, demonstrating how far off the mark the idea that experiments are merely the passive observations of an operation that runs on its own is. Hacking and Latour have more extensive elaborations, which are thus harder to summarize. But one of the main issues in Hacking's paper is the taxonomy of laboratory items that he proposes: things, ideas and marks - and how scientists operate on them in order to produce the "best fit" (his words) among the various items involved in an experiment. Latour stresses the lab experiments as a means of producing, through very simplified models (a scientist that is dealing with culture plates has far less in his/her hands than a farmer watching over cows in the field) quick cycles of trial and error, so that a lot of learning and mastery can be gained from this series of trials and errors. Both underscore the fact that science quickly became concerned with phenomena that it produces, instead of simply registering "observations" passively - a point that echoes a concept proposed by Gaston Bachelard - phenomenotechnique (that is, the extent that phenomena are created and theories built in instruments). Both also note how science, once having produced stable artifacts and statments in the laboratory, "colonizes" (my expression) the world at large, through the imposition of standards (Latour uses the specific example of metrology) and other technical interventions.
Taken together, these texts provide a compelling - and documented - case for what became known as the "theory-ladeness" of observation, which is yet another connection to my earlier post on objects and ideas. They also share other common ideas, more or less emphasized on each individual paper, especially the collective character of the knowing subject (rather than the individual scientist) and the fundamental role that practical skills play in the development of laboratory apparatus and, because of that, whole theories.
This has important implications for medicine, particularly if we take into consideration how much closer caring for patients in wards is to a "laboratory experiment" (in terms of a simplified environment fully under control) than, for instance, outpatient care. It would thus follow that such settings - and particularly the more high-tech ones, such as ICUs and ORs - boost tremendously the impression of effectiveness of biomedicine, since we are dealing with isolated episodes in limited periods. When the life cycle of the patient is considered, though, the effective "control" seems a lot less impressive.
There are obviously much more implications, but for the moment I'll leave those as an exercise for the reader.

References:
- Fleck L The genesis and development of a scientific fact. Chicago:The University of Chicago Press, 1979
- Gooding D "Putting agency back into experiment" in Pickering, A. (ed) Science as practice and culture. Chicago:The University of Chicago Press, 1992
- Hacking I "The self-vindication of the laboratory sciences" in Pickering, A. (ed) Science as practice and culture. Chicago:The University of Chicago Press, 1992
- Latour B "Give me a laboratory and I will raise the world" in Biagioli M The science studies reader. New York:Routledge, 1999

Bernal

Every semester I have a running seminar on some aspect of science studies. This semester we are going through a series of texts of sociology and anthropology of science/knowledge (there are some important distinctions but I won't bother with them in this post). Today we discussed parts of the classic The Social Function of Science by JD Bernal. I added this reference to the bibliography of the seminar given its enormous historical importance as a (maybe the) seminal text of the whole field of science studies.
Bernal was an extraordinary figure. A researcher that pioneered X-ray crystallography, was part of the British war effort, helping plan the D-day invasion of Normandy and being one of the people responsible for the creation of the floating harbours ("mulberries") essential to the establishment of the beachheads there, political activist and a pioneer in sociology of science (who influenced Merton, who in turn influenced Kuhn).
Having said that, and not meaning in any way to belittle the man, his text didn't age that well. He has some remarkable insights, particularly at one point at the end of the book where he seems to be anticipating the idea of complexity and emerging properties, which became so important in recent years (another post for another day?). But his naïve belief in science as the force that would drive mankind to a better life, coupled with his acritical defense of the Soviet science is almost embarassing to read. We must cut him some slack, anyway; the book was published in the late thirties, before the onset of WWII, before the Ribbentrop-Molotov pact, before Stalin's crimes where known outside the USSR, and at a time where soviet-style communism seemed the only alternative to nazi-fascism. And particularly, as far as the biomedical sciences are concerned, before the Lyssenko fiasco was exposed to the world.
I read his text as a cautionary note against tying scientific endeavours too closely to political-ideological systems, even when we are sympathetic to them. If there is one point to be made here for me it is that although ideas such as scientific neutrality, impartiality and objectivity can be demonstrably wrong and more or less easily brushed aside, they nevertheless may constitute an important set of regulating values for scientific practice, values which practitioners will strive to demonstrate their committment to (even if just for external consumption) because other practitioners - their colaborators/competitors/consumers/referees, as Bourdieu stressed on his paper on the scientific field - will keep them in check.
During the seminar this developed into a lively discussion of reason, science and the Illuminist tradition, but that will wait for another day.
More on Bernal: http://www.comms.dcu.ie/sheehanh/bernal.htm

Objects and Ideas

One of the most recurrent criticisms of "social constructionism" (this label will probably have a post dedicated to it further down the road) that I am aware of goes more or less along the following argument: it only makes sense to talk about the social construction of the ideas that we have about things, but not of the things themselves. But what access do we have to "things" that does not include "ideas"? What "epistemological knife" allows us to separate the latter pair?

I think that one of the most interesting discussions of this point (again: that I am aware of) is Ludwik Flek's take on anatomy, in his "The genesis and development of a scientific fact". Flek presents Vesalius' anatomic atlas to point out the role that styles of thought (one of his key concepts) bear on our descriptions of the world. Since in Vesalius time it was assumed that the sexual organs of men and women were fundamentally homologous, the illustrations of the female organs included a duct that would carry the "seed" to the womb - a female counterpart of the ductus deferens. Flek then goes on to say that he wanted to contrast this depiction with modern illustrations that would show actual anatomic facts; but it turns out that anatomic atlases would present a stylized view of the human bodies. He describes a dissection manual that had photographies, but those in turn were snapped along favourable axes, included lines and arrows to point out certain structures and so on and so forth. Flek concludes, then, that it is impossible to compare theories with facts; in the end, the comparison boils down to an old theory versus a newer theory.

Does this mean that Flek spoused an idealist position? Not quite. He states that theories have arbitrary connections that are the result of the researcher's decisions, which he called active, but others that were due to the objects being researched, which he called passive, that is, independent of human agency. He gives atomic weights as an example: that the weight of hydrogen is (approximately) one is arbitrary, but once you set that, the atomic weight of oxygen has to be (approximately) sixteen. And then he goes on to state that as a discipline progresses (my word), the number of passive connections increase - but so do the active ones.

So this brings me back to the beginning of this post. "Ideas", or active connections, are as integral to scientific objects as the "things", or passive connections, related to them. True, in the artificially simple example above it's easy to tell which is which, but considering the plethora of connections that any given scientific theory has, the entanglement of active and passive connections makes it impossible to establish, once and for all, where "ideas" stop and "things" begin.

To wrap it up, this is why I do not find particularly compelling a critique that hinges on this a priori separation. Your thoughts?