Torture: is it ever justified to inflict pain to extract information in extreme cases?

Jeremy Bentham, answered from the texts and cited to the page.

The position in brief

In early manuscripts, written from the legislator's point of view before I had set out the law of evidence, I defined torture narrowly — violent pain applied to compel a man to do or forbear, ceasing the instant he complies — and distinguished it sharply from punishment, which it must never be. So defined, it deserves less disapproval than punishment, and under strict conditions (its purpose clear, its application ceasing at once, the thing required known to be in the man's power, the same proof as for punishment, the public interest served, the kind and degree regulated, compensation to any wrongly used) I allowed it, the threshold falling as the danger to the public rose.

From the later point of view of the people, having written out the law of evidence, I came to hold that cross-examination, the non-exclusion of relevant evidence, and a complete utilitarian code furnish what is wanted without it.¹²

Sources

1. For rendering it such, keep in mind this radical allusion. The commun¬ ity is the body politic. Misdeeds are it's diseases. Occupied on the Penal Branch of Law, the Legislator is its medical practitioner—its surgeon. In a surgical operation, the cure is the benefit: the pain of the patient, the burthen.

   Legislator of the World, Writings on Codification, Law, and Education, p. 209

2. His case had been publicized by Francois Marie Arouet, usually known as Voltaire (1694-1778), who in 1765 succeeded in having a posthumous pardon granted to Calas. Damiens was sentenced to be pulled apart by four horses; however, even six horses failed to accomplish the feat, and eventually the executioners had to cut the limbs away from the body.

   Legislator of the World, Writings on Codification, Law, and Education, p. 77

Where the argument is made

• The two manuscript sequences on torture (written c. 1770s-1780s, part of the material for a penal code, not published by Bentham and awaiting authoritative edition in the Collected Works)
• the later writings on judicial evidence and procedure (Rationale of Judicial Evidence
• Of the Limits of the Penal Branch of Jurisprudence, ed. Schofield 2010)

How the argument was received

The early position survives only in unpublished manuscript fragments (Schofield), framed from the legislator's standpoint of the 1770s-1780s. The presumption is that Bentham ceased to recommend torture when he returned to judicial procedure after 1803 and rewrote it from the standpoint of the people. The mature writings on evidence supersede the early manuscripts; the full reconstruction awaits the Collected Works.