Problems with the Biomedical Model of Health

AAP Conference, Adelaide. 2003.


Defining the concept of ‘health’ is a central problem in the field of Bioethics. It sits at the crossroads of multiple levels of theory, having implications for both metaphysics and normative ethics. Among the plethora of potential theories, the work of Christopher Boorse continues to be widely influential. Boorse argues that it is possible to give an objective, value-neutral account of the concept via the device of teleological function statements. Although people may value the states that fall under his description, it is no part of his theory that body states are healthy because they are valued. Contrary to Boorse, I will argue that his choice of teleological goals cannot be derived merely from observation and that they are in fact value-laden in ways that are unlikely to generate universal assent as a plausible account of health.

Under Boorse’s model the “state of an organism is theoretically healthy, i.e., free from disease, in so far as its mode of functioning conforms to the natural design of that kind of organism”. Diseases, including injuries and congenital disabilities, are “deviation[s] from the natural functional organisation of the species”. The concepts of ‘biological or natural design’ and ‘natural function’ are central and inter-linked in this model of health but their interpretation, is not self-evident. We cannot for instance simply interpret ‘natural’ functioning as a statistical average of human functioning. As Boorse himself observes, if we take this approach both too much and too little is included under the concepts of disease and health. If health is a statistical average, then an average life span of forty-eight years was healthy before nineteen hundred, but an absence of tooth decay now is not. Similarly, individuals with type O blood will not be simply unusual but diseased. That these implications are strongly counterintuitive, might in itself be a basis for rejecting the ‘statistical average’ interpretation of the model. However, a more serious objection is the apparent lack of rationale to statistical averaging.

Statistical averaging takes the mere contingent fact of current statistical averages as the basis for our concept of health but provides us with no obvious reason for doing so. The statistical averages of human biology have changed significantly over time and we presume will continue to do so. A host of variables including average height, weight, lifespan, and so on have all altered significantly in the past century, both across the human population and within subgroups of the population. Statistical averaging as an interpretation of ‘normal functioning’ provides no obvious reason to favour current averages over either past or future averages. There may of course be ‘reasons’ for favouring current average life spans for instance, over past averages. However, these ‘reasons’ arise from views of human wellbeing that are external to and independent of, any view of statistical normality. In other words, although a statistical average is itself value neutral, any reason we may have for favouring one average over another, involves explicit value judgments and this is precisely what Boorse hopes to avoid. Boorse himself concedes that, by itself, statistical averaging is an insufficient basis for a theory of health and an unconvincing interpretation of ‘natural’ functioning.

Instead he opts for a teleological interpretation of function and design that, while containing the concept of statistical averaging as an important component, does not rely solely on it. Under this interpretation, the design of an organism or part of an organism is determined by its actual functional role and a “function in a biologist’s sense is nothing but a standard causal contribution to a goal actually pursued by the organism.” The biology of living organisms is therefore a series of ascending or descending functional goals. The ‘design’ of the heart, for example, is determined by the functional goal it typically fulfils in humans. It pumps blood around the body in a manner typical of human hearts and consistent with higher level functional goals. Thus, while the immediate functional goal of the heart is to pump blood, pumping blood serves the higher level functional goal of “individual survival and reproduction.” A healthy heart is analogous in this sense to a functional carburettor. Both are ‘designed’ to serve functional goals within larger functional operations and both are functional if they actually serve those goals. According to Boorse, health under this model is “strictly analogous to the mechanical condition of an artifact.” Human beings are healthy if their various biological components operate in a manner that is both species typical, i.e., statistically typical, and consistent with their higher level biological goals of reproduction and survival.

However, there are a number of difficulties associated with the functional model, that arise from the absolute distinction between fact and value that the model insists upon. Boorse himself observes, under functionalist models “health and disease belong to a family of typological and teleological notions which are usually associated with Aristotelian biology and viewed with suspicion.” In particular, Boorse in keen to dissociate the functional concept of ‘natural design’ from the “view that an ideal can be simultaneously empirical and normative”. He argues that the empirical status of an organism’s natural design tells us nothing about the normative value of such a design nor of any functioning associated with that design. Health under this model is strictly an “instrumental good rather than an intrinsic good” It has value in as much as it enables us to achieve other ends that may or may not be morally desirable but it has no intrinsic value of its own.

The instrumental value of health that this model envisages is most clearly seen in lower level functions. Lower level functions only have instrumental value in relation to the higher level functional goals. A heart that does not serve the goal of survival usually has no instrumental value to its possessor. Certainly, whether it functions or not, the human heart lacks intrinsic value. Provided that an artificial heart performs precisely the same functional operations as a human heart, with precisely the same efficacy, the functional model provides us with no reason to prefer one over the other just as we have no reason to value one functional carburetor over another. Equally, it is no part of this description that we should value the ends to which health is an instrumental means, any more than we should value a functional carburetor when we have no use for a car. As Boorse argues “it is no part of biological theory to assume that what is natural is desirable,” nor the reverse.

While there is a substantial body of literature dealing with critical analysis of Boorse’s work, the most interesting recent assessment has been by Scott De Vito De Vito argues; persuasively; that Boorse’s project of developing a ‘value-neutral’ description of ‘health’ is doomed to failure. In the very act of choosing physiology as a basis for establishing criteria for ‘health’ Boorse makes an evaluative judgment about which criteria are important. He chooses physiology over other possibilities, for instance genetics, and he chooses the interests of physiologists over the interests of other potential stakeholders, for instance patients. As De Vito points out, there is nothing in Boorse’s theory that obliges us to accept either that physiology is the basis of medicine or that “medicine does, should, or must use the physiologists view point in defining health and disease.” At the very first level of the theory, the choice to adopt the criteria of physiology is value-laden. However, it is not the choice of physiology that is of most concern here. Rather, it is the value-laden nature of criteria chosen under the framework of physiology that is of concern, in particular the choice of ultimate teleological goals.

The ultimate teleological goals chosen by Boorse, the goals of survival and reproduction, are particularly problematic. While he is prepared to concede that the choice of teleological goals, to some extent reflects the interests and therefore the values of biologists, nonetheless he believes that the choice of goals is supported by observation and that the goals themselves are not value-laden. Neither claim seems to be justified. If individual survival is the higher order goal that all organisms actually pursue and against which we are to judge functional health, then aging is ipso facto a disease. Mortality is the ultimate functional aberration. While it may appear self-evident that all livings things age and eventually die, this is inconsistent with the claim that it is a matter of empirical observation that all living things are designed for survival.

Certainly, Boorse does not wish to view aging as a disease or disfunction, rather he argues that aging is an “inherent defect of the organism”. However, his reasoning on this is less than persuasive. If aging is an inherent feature of our functional design as a species, we have no reason to describe it as a “defect” unless we already have some view of our teleological goals that is not supported by observation. If, on the other hand, aging is a “defect” then we there is no reason why we should not to describe it as a disease or disfunction. Given the essentially mechanistic view of human physiology that the functional model projects, it is perhaps unsurprising that a form of immortality should be the inevitable consequence. Provided that we continue to replace worn out parts, a T-Model Ford will continue to run indefinitely. Under Boorse’s model, this is also apparently the case with human beings. In principle, provided we continue to replace or repair worn out parts, we too will continue to function indefinitely.

However, if we are inclined to doubt that functional immortality is the goal of living organisms and that our biological design serves that end, then the functional model of health faces a serious problem. If we are not designed for indefinite survival then the description of this as a central functional goal is prima facie false. At the very least, the description of our central functional goal would need to be modified to suggest that we are functionally ‘designed’ to survive as individuals not less than X number of years but not indefinitely. While it is possible that we may arrive at some view of our species typical life-span, such a view would seem to require further substantial modifications to the concept of teleological design. A fixed, species typical life span would reintroduce the problem of defining our functional goals and, in this instance, it is not obvious what our functional goals would be. We might describe our higher level functional goal as built-in obsolescence rather than survival. There is no obvious reason that can be derived from observation why we should prefer one description to the other.

That Boorse chooses individual survival over built-in obsolescence as the goal of biological functioning, reflects a quite specific view about what is valuable and this is not a view that is invariably compelling. It might be argued that, other things being equal, people do typically value individual survival. Some assumption like this seems to be why Boorse has chosen survival as a functional goal. However, we are not forced to this view. We might with equal plausibility value a limited life span as a teleological goal. Certainly, there are clear cases where individuals do not value continued survival. It is also interesting to note that a mechanistic view of health, that views human functioning as sustainable through the continual replacement of worn out parts, constantly in pursuit of individual survival, has given rise to some of the more intractable problems in Bioethics.

Similar difficulties arise with Boorse’s choice of reproduction as an ultimate teleological goal. Once again Boorse’s claim that such goals are the goals “actually pursued by the organism” runs into counter examples. Many species simply do not reproduce at the level of individuals. For instance, ants, bees, and mole rats all contain a majority of individuals who are reproductively sterile. Reproduction is simply not a goal actually pursued by the majority of individuals in these species. Even if we restrict our interest to humans, there seem to be clear counter examples. Lactation is an effective form of fertility suppressant in humans. Women who are lactating are effectively infertile. Under Boorse’s model they would also seem to be suffering a failure of functioning and are therefore strictly speaking, unhealthy. Equally, post-menopausal women are infertile, under Boorse’s model it is not at all clear what their functional organization is ‘designed’ for. They too seem to be by definition, unhealthy.

These examples also illustrate two other interesting aspects in the choice of teleological goals. Firstly, Boorse seems to have failed to observe his choice of teleological goals may have different ramifications between the sexes and secondly, that the goals themselves may come into conflict. That the goals of reproduction and survival sometimes come into conflict is easily observable in other species where the death of individuals is either a precursor or inevitable consequence of reproduction. However, the tension between reproduction and survival is also observable in humans. Not only are there particular instances where the goals of reproduction and survival come into conflict in the bodies of individual women, but depending on how one understands the teleological goal, it may appear to be more generally in conflict with women’s survival. If the goal of reproduction is interpreted as unconstrained fecundity; and this seems to be how a number of theorists using Boorse’s model have interpreted it; then it seems an unlikely choice of teleological goal. Simply, too many children, too close together impairs other aspects of women’s biological functioning. Ultimately, it adversely impacts on their teleological goal of survival. Since the teleological goals nominated under Boorse’s model of health are not sequentially ordered, any conflict between the two goals invariably involves an evaluative choice between goals.

I hesitate to suggest that there is a gender bias in Boorse’s choice of teleological goals yet it is difficult to avoid the suspicion that it is a male model of human physiology that has been taken as the basic description of biological functioning. Typically, the organizational goals of reproduction and survival will not come into conflict in males and, post puberty, males are either fertile or they are not. Not only does reproduction always come with a survival risk for women but the majority of women will continue to live long after reproduction is a plausible biological possibility. Certainly, it is not clear that ‘reproduction’ as unconstrained fertility, is a plausible choice of teleological goal for women’s biology.

Boorse’s contention that our ultimate functional goals are a relatively straight-forward matter of empirical observation is ultimately unpersuasive. The teleological goals in particular seem to have been chosen precisely because it has been assumed that these are goods that, other things being equal, people value. However, if it is the case that Boorse’s account of the concept of health is value-laden then its plausibility is seriously undermined. Clearly, people value many states of functioning and states of being other than those of survival and reproduction. Equally, people do not always value either reproduction or continued survival. As a value-laden concept, Boorse’s choice of teleological goals seems arbitrary.


Boorse, Christopher. “Wright on Functions”. The Philosophical Review . LXXXII, January 1976. 70-86.
“Health as a theoretical concept”. Philosophy of Science . Vol.44. 1977. 547-573.
“On the distinction between disease and illness”. In Marshall Cohen, Thomas Nagel and Thomas Scanlon. (eds) Medicine and Moral Philosophy. Princeton: Princeton University Press. 1981. 49-68.
Dawkins, Richard. The Selfish Gene . Oxford: Oxford University Press 1976.
Daniels, Norman. “Health-care needs and distributive justice”. In Cohen, Marshall., Thomas Nagel and Thomas Scanton. (ed) Medicine and Moral Philosophy . Princeton, Princeton University Press. 1981. 81-114.
De Vito, Scott. “On the Value-Neutrality of the Concepts of Health and Disease: Unto the Breach Again.” Journal of Medicine and Philosophy . Vol., 25, No. 5. 2000. 539-567.
Lennox., James. G. “Health as an objective value.” p.500. The Journal of Medicine and Philosophy . Vol.20, No.5. Oct. 1995. 499-511.
Sade, Robert.M. “A theory of health and disease: The objectivist-subjectivist dichotomy.” The Journal of Medicine and Philsophy . Vol. 20, No.5. Oct. 1995. 513-525.
Wright, Larry. “Functions” The Philosophical Review . LXXXII, April 1973. Pp.139-168.