The same could be said of making your staff drink honey when they come into work. If it's not increasing their productivity, perhaps they're simply not drinking

The same could be said of making your staff drink honey when they come into work. If it's not increasing their productivity, perhaps they're simply not drinking enough of it. It's true enough that nothing will work if you don't sufficiently commit to it. The limitation with the observation is that it doesn't show how well associated different stimuli are with the desired output. Did torture not get the truth, perhaps you just didn't use enough of it? Did dosing someone with Omeprazole not cure a cold? Perhaps you didn't use enough of it... and so on. Or perhaps it's not the right approach. What we can say in general terms is that if you need a lot of something, the desired effect is at least only weakly associated with the stimuli. It might be possible to get the desired effect by using more of the thing, but this is a poor substitute for understanding the mechanism involved, and can be prohibitively costly; ethically and practically. In the given case: being an arsehole, for all that it comes easily to the individual, seems likely to be expensive to those around them - and by extension to the organisation - in other terms. There are people who will refuse to work with you because you're an arsehole. There are people who will burn out if you push them too hard. There are people who will keep their own counsel on issues rather than attempting to involve you in things. There are normative arguments that you won't be able to make, because no-one will take them seriously coming from someone like that. Balanced against the uncertainty that the output is even there, (anecdotes seem to go both ways on the issue,) that it does not match how I would expect productive working relationships to operate, and that I personally have never seen an effective team that I would attribute to the manager being mean, (and a great many ineffective teams that I would attribute to the same,) it does not seem a wise strategy to me. Though, that said, you might be bringing a different model of productive working relationships, or a different basis of observation, to the discussion. (Or be playing devils advocate.)