The Interdisciplinary Fellowship in Health Care program has brought scholars from a variety of backgrounds to the Institute of Human Values in Medicine at the Medical University of South Carolina. My own background is in linguistics, the field that defines itself as the study of language, and one of my chief research interests is in medical communication. In this article, I will briefly summarize disciplinary linguistic research on medical communication, and then describe the interdisciplinary perspective I am developing as a result of the Fellowship program.

The field of linguistics describes the ways that language is structured and the ways it is used in different contexts. To a linguist, language is both a cognitive and a social object. Language is a deeply internalized competence, a set of rules for the production and interpretation of new sentences. The rules of language encompass the sound system, the word system, the grammatical system, and the meaning system of a language: English, for example, has sounds like /k/ and /l/, words and parts of words like table and the plural {-s}, and sentences formed in subject-verb-object patterns. The rules of language also encompass some aspects of meaning: for example, the difference in meaning between John is eager to please and John is easy to please comes from the different meanings of the words eager and easy as well as the grammar of the sentences—John is the subject of eager, but the object of easy. But meaning is only partially described in cognitive terms. Much more of meaning derives from language used in different social contexts, and meaning in context is a complex matter of conventions rather than rules. Take, for example, naming conventions. What does it mean when two individuals address each other by first names? Usually it means a symmetrical social relationship in an informal situation. Similarly, what does it mean when one individual address another by title plus last name while that individual responds by using a first name? Usually it means an asymmetrical social relationship in an institutional situation; a medical encounter, for example, or a medical team with an attending physician in charge of housestaff fellows, residents, and interns. But using names in context is much more complex, and these basic rules often do not apply straightforwardly. In hospitals today, housestaff are trained to call patients by Mr./Mrs. plus last name, although the symmetrical naming does not change the basic institutional situation of a professional and a client. On rounding teams, attendings and fellows often use first names reciprocally, but this seeming symmetry is in name only, since there remains an asymmetrical relationship between the attending responsible for the work and training of the fellow. Language in context, then, is both patterned and particular, the same and different, the unique and the mundane.

So language as a social object is enormously complex, and the way that linguists look at language in context is more descriptive and qualitative than formal and quantitative. Many linguists looking at language in context use a method called discourse analysis, which is an attempt to identify important features and conventions of language in use and to provide for them what the anthropologist Clifford Geertz calls a "thick description," an explanation of the way that a convention of language points to meaning and interpretation. A thick description of naming in a hospital, for example, would collect multiple examples of naming practices, compare them to one another, and develop a description of the different meanings of names in a hospital setting, paying special attention to naming practices that raise troublesome social issues such as false respect or problematic authority.

Using this qualitative and descriptive perspective, the field of linguistics has investigated the language of medicine. Early research focused on the ways that medical discourse reflected an asymmetrical institutional context, with physicians in control of the discourse. Richard Frankel (1979), for instance, showed that physicians typically interrupted patients less than 15 seconds into the beginning of a medical appointment. Candace West (1984) found that patient questions comprise a very small part of physician-patient discourse, only 9% of all utterances. Sue Fisher (1986) discovered a disturbing finding in a comparison of private practice and social clinic medicine: women with abnormal Pap smears in a private practice setting were offered a variety of treatment options, but in a low-income clinic, patients were offered the single option of hysterectomy, with other options only discussed if the patient specifically asked about them. Given that patients can have a hard time claiming the floor for questions or discussions, as West and Frankel have shown, this kind of medical communication can be detrimental to good medical care and physician-patient relationships based on respect and integrity.

More recent research on medical discourse has looked at the ways that physician-patient communication is interactive in very complex ways. Paul ten Have (1989), for example, looked at the typical structure of a medical encounter (opening, complaint, examination, diagnosis, treatment advice, closing) and argued that both patients and physicians contribute to the smooth flow of an appointment. Doug Maynard (1992) observed that even sequences that would seem to "belong" to the physician, so to speak, can actually be interactively co-constructed by physician and patient jointly: in diagnosis, for example, physicians often elicit patient's views of their conditions, or the conditions of their family members, with questions like How do you see (Billy) now? Christian Heath (1992) found one way that patients solve interactional problems: if patients have unanswered questions or unresolved disagreements, they bring up their concerns during the closing portion of the question, not the portion of the encounter in which the physician asks whether patients have any questions.

My own research on medical communication also looks at the co-construction of specialized sequences. In a paper on the ways physicians make referrals with families who have children with disabilities (Barton, in press), I looked at the ways that expertise and compliance are two intertwined dimensions of the institutional context of a medical encounter. The argument of the piece is that families who display their expertise in and compliance with the medical system have markedly different referral and account sequences than families who do not display their expertise and compliance. Let me briefly show you some different referral sequences -- (1a) is a referral with an explanation for a family new to the system of pediatric specialty care, and (1b) is a second referral still with an explanation but with a non-compliant family:

Example 1(a):
Mom: "I have a concern that I’ve mentioned two or three times but no one seems to think it’s a big deal. But it’s a big deal to me. ((Amy)) has a problem with moving her bowels. To the point that she is bleeding and crying hysterical"

Dr. G: We’ll stop the "oh mother" situation. We’ll send her to GI because what can happen is that the rectum has a problem. The large bowel may not have the nerve endings that it needs. So it may not pass, so it accumulates. It can cause fissures and bleeding and pain. So the question is what is the problem? I’m not a GI person, but I’ll refer her to ((Pediatric Hospital)) and they have a dietician with them so they can also refer the proper diet.

Continued ...