While posting this essay, I found several typographical errors that I somehow managed to miss while proofreading it for submission. I've corrected them and added some HTML links to provide explanation and further reference. This said, my essay is as follows:
This is a pretty difficult question for me to answer, mostly because I believe the question itself is both flawed and loaded. To begin with, our textbook grossly misuses the term "abnormal", equating abnormality with dysfunction and defining the term "abnormal behavior" as synonymous with the concept of a psychological disorder (cf. p. 2). In actuality, however, the term "abnormal" simply means "not normal" and is synonymous with the term "atypical". As such, any behavior which differs from the norm is abnormal… and atypicality is the only criterion with any relevance.
While it is possible to argue that our book uses "abnormal" and its derivatives in a special sense, the use of this term in such a fashion is a gross violation of both medial bioethics and clinical ethics in general. Specifically, it violates both the principles of benevolence and non-maleficence through its effects both on informed persons who accept this equation (i.e. students learning from the book) and the effects of regarding this sort of linguistic equivocation as acceptable on society as a whole.
I certainly do not believe in linguistic determinism, but it is very difficult to deny the existence of a relationship between language and thought. How we think affects the sort of language we use, and there is a good bit of evidence that the sort of language we use effects the way we think (e.g. Tan et al., 2008). The existence of several named logical fallacies related to this (e.g. equivocation, amphiboly, fallacy by semantic shift) tends to support this general idea. Given the state of the knowledge in this arena and the precautionary principle, there is an ethical obligation to avoid this sort of issue.
Moreover, there is the matter of laypersons who hear discussions using this sort of misused terminology. It is unlikely that such persons will be aware of the issues surrounding the constructions used, and we live in an era in which the technical literature is becoming increasingly available to such people (as evidenced by publications such as PLoS and the policies of PubMed Central).
Even if the issue of the false equivalency created by this abuse of the English language is ignored, the criteria laid out in the book are problematic on other grounds. Simply put, they have been used to justify outright bigotry and abusive "treatments" which were only "justified" by this bigotry (e.g. Cartwright, 1851; Lovaas, 1987; O'Malley, 1914; Rekers & Lovaas, 1974; Chapter 14 of our book). The use of our book's criteria as sole and sufficient demarcation for what is and isn't a mental disorder is flawed, highly unethical, and not what is actually done in practice.
Furthermore, the entire construct of "mental illness" is of dubious validity and has been extensively challenged (e.g. Szasz, 2008). Even if we can't demonstrate that mental illness objectively exists, however, there remains one potential justification for the use of the concept: utility.
Simply put, we can justify describing something as a mental illness if we can demonstrate that the concept – and the medical-model approach which goes along with it – is useful. Doing so in a clinical environment, however, would require a demonstration of improved patient outcome, and it is very difficult to conduct a rigorous randomized controlled trial of the use of a concept.
Despite this, it is possible to evaluate the consequences of medicalization via the epidemiological evidence and comparing the prognosis of various syndromes in countries that have widespread access to medical treatment for mental illness and those that do not. It is also possible to do this by comparing outcomes across time-periods in a similar fashion.
Scizophrenia is the most "extreme" of the DSM mental disorders, and it is a major mainstay of psychiatry. As such, it makes intuitive sense to begin by evaluating the utility of the concept of a psychological disorder in reference to schizophrenia. Moreover, there exists a single randomized, controlled trial of psychiatry as a whole in the treatment of schizophrenia: the Soteria Project.
None of these approaches yield evidence in favor of the utility of a medical model approach to schizophrenia: World Health Organization data indicates that schizophrenia has better prognosis in developing countries than in developed countries; the best outcomes to be found for the treatment of schizophrenia here in America can be found in those non-medically treated by Quakers in the 19th century; and the findings of the Soteria Project were very much damning in regards to the effects of psychiatric intervention (see Whitaker, 2002 for review).
Or, in other words, the concept has yet to demonstrate utility in regards to treatment. If and when it does, I'll start to consider medical-model approaches ethically justified.
References
Cartwright, S. (1851). Report on the diseases and physical peculiarities of the negro race. New Orleans Medical and Surgical Journal, 7, 691-715.
Lovaas, O. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9.
O'Malley, M. (1914). Psychoses in the colored race. American Journal of Insanity, 71, 309-337.
Rekers, G. & Lovaas, O. (1974). Behavioral treatment of deviant sex-role behaviors in a male child. Journal of Applied Behavior Analysis, 7, 173-190.
Szasz, T. (2008). Psychiatry: The Science of Lies. Syracuse, NY: Syracuse University Press.
Tan, L., Chan, A., Kay, P., Khong, P., Yip, L., & Luke, K. (2008). Language affects patterns of brain activation associated with perceptual decision. Proceedings of the National Academy of Sciences, 105, 4004-4009.
Whitaker, R. (2002). Mad in America: Bad Science, Bad Medicine, and the Enduring Mistreatment of the Mentally Ill. Cambridge, MA: Perseus Publishing.
Edit: Corrected one more typo...