Debate: Are mechanisms required to establish treatment effects?

27/03/2017 @ 6:00 pm – 7:30 pm
Oxford University
Rewley House
1 Wellington Square, Oxford OX1 2JA
Debate Organiser

An enduring debate in the history of medicine exists between people who believe that we need to know how a treatment works in order to know that it works and those who believe that careful observation is enough.

Celus (25 BC—50 CE) reports differences between ‘Empirics’ (who held that careful observations are enough), and ‘Dogmatics’ (who insisted that we need to understand the underlying causes and mechanisms). [3] Centuries later Roger Bacon (c.1266 CE) de-emphasized the role of understanding mechanisms and causes, while Descartes (1596 CE—1650 CE) believed that things in the world—including humans—are machines so we need to understand mechanisms to diagnose and treat patients.

More recently, a ‘new mechanical philosophy’ has emerged, with some philosophers of science arguing that unless we have evidence of a mechanism that a treatment works, we do not know whether it works. [1] This leads them to assert, for example, that Semmelweis did not know washing hands saved lives although he had strong experimental data, because he did not propose an acceptable mechanism. This philosophical view is opposed to a view commonly held by proponents of Evidence-Based Medicine (EBM) who point out many historical cases ranging from lemons to cure scurvy to aspirin for reducing cancer incidence where we have no idea what the mechanism is yet we believe we know the treatments work. [2]

To be sure, proponents of mechanisms are not against careful observations, they just don’t believe these are enough. Likewise, proponents of EBM do not deny that there are mechanisms or that they can be important, they just don’t think mechanistic evidence is required.

In an epic attempt to resolve this long-standing debate once and for all, leading contemporary proponents of each view will make their cases.

  • Professor Jon Williamson will argue that we do need to have evidence of mechanisms in order to prove treatment works.
  • Dr Jeremy Howick will argue that while evidence of mechanisms can be useful, they are not required to establish that a treatment works.

Moderated by Professor Jeffrey Aronson.

This debate is being held as part of the History and Philosophy of Evidence-Based Health Care module which is part of the MSc in Evidence-Based Health Care.

This is a free event and members of the public are welcome.


  1. Russo F, Williamson J: Interpreting causality in the health sciences. International Studies in the Philosophy of Science2007, 21(2):1157-1170.
  2. Howick J: The Philosophy of Evidence-Based Medicine. Oxford: Wiley-Blackwell; 2011.
  3. Howick J: Aulus Cornelius Celsus and ‘empirical’ and ‘dogmatic’ medicine. JLL Bulletin: Commentaries on the history of treatment evaluation 2016.

Leave a Reply

Your email address will not be published. Required fields are marked *