Allow me to start this editorial with question. A patient with DSM-IV diagnosis of major depression (or dysthymia) is sitting in front of you. What does the diagnosis tell you about its causes and management strategies? Little of value, I would suggest. A diagnostic system's utility is limited when it advances descriptive categories that are remote from causality, have limited validity and reliability, and fail to provide management options. The DSM model has had its uses, but in light of present knowledge, it is now time to contemplate an update. As most journal readers received their psychiatric education following the introduction of the seminal DSM-III model in 1980, any challenge to this system may appear discomforting and this editorial unnecessarily provocative. The DSM's relevance to most clinicians is high, at least to those who recognize its potential to meet small set of quite practical needs; its lexicon offers ready communication to colleagues and many patients, its status aids the writing of medicolegal reports, and the gravitas of some diagnoses (for example, major depression) assists patients' access to insurance and hospitalization. Its designers, however, were more aspirational. In the introduction to DSM-IV, the Task Force states that the utility and credibility of DSM-IV require that it focus on its clinical, research, and educational purposes and be supported by an extensive empirical foundation. Our highest priority has been to provide helpful guide to clinical practice . . . An additional goal was to facilitate . . . More than any other nomenclature of mental disorders, DSM-IV is grounded in empirical evidence (p xv-xvi). In this issue of The Canadian Journal of Psychiatry, the utility of the DSM classificatory system for depressive disorders is considered from polarized positions. Being aware of Robert Goldney's support for the system and his capacity to marshal cogent points, I invited him to put forth the argument for maintaining the current DSM model, while I provide an opposing critique. Goldney argues that the DSM operational definitions of broad depressive syndromes, allows a comprehensive description of each depressive episode, and has proven to be ... as an international tool for psychiatric research (1, p 876). In my opposing piece (2), I note intrinsic logical limitations to any strategy that dimensionalizes categories and categorizes dimensional constructs. I suggest that, in practice, DSM categories are poorly defined, its criteria sets are problematic (compromising reliability estimates), and it lacks utility or explanatory power in terms of informing us about causes or treatments. I further suggest that these consequences are hardly surprising when constructs such as major depression are reified without challenge. Such charges proceed beyond being polemical-they are either supported or rejected by evidence. This journal is to be congratulated on initiating such debate, as it allows the parameters to be defined and challenges us to engage further with such issues. Goldney is provoked by my writing style on such matters. While the issues should be debated on their substance, in relation to style, I plead the Ricky Gervais defense: you're not offending someone, you're not doing anything. As colleague once observed, collegiality can be compromising, and every organization should have prickly bastard who keeps asking questions. The question that I raise is: If the DSM provides model for the depressive disorders that compromises and clinical practice, then, and invoking another observation, what's the use of running if we're on the wrong road? Certainly, as Goldney argues, the DSM is influential in offering tool for psychiatric research. Most reports respect the DSM model and describe depressive samples as meeting DSM criteria for major depression or dysthymia. However, being influential does not necessarily imply or demonstrate utility. …