“Cognitive Enhancement: Defending the Parity Principle”, St Cross Special Ethics Seminar by Neil Levy

Last Thursday Professor Neil Levy has defended his Parity Principle for analysing the ethics of cognitive enhancement at the St Cross Special Ethics Seminar. Such principle would oppose a common form of objection against enhancement which claims that there is a worrying asymmetry between enhancement and traditional means to human improvement. Conversely, Neil contends that the function is all that matters morally when comparing enhancement with traditional means and that comparing isofunctional modifications reveals that there are little unique problems with enhancement. The Parity Principle leads to a useful analysis of several proposed critiques of cognitive enhancement.

Firstly, there are safety arguments against cognitive enhancement, which appear to be dismissed by recent empirical evidence and, moreover, of little philosophical interest. Secondly, there are authenticity concerns relating to the fact cognitive enhancement implies either the imposition of an exterior disposition or a mismatch with reality. But a given traditional mean to achieve a functionally similar result to a given enhancement seems to imply these same two problems. For instance, both Cognitive Behavioural Therapy and SSRIs for treating depression can be equally said to be an exterior disposition or lead to a mismatch between one’s emotion and reality – he seems to find these unproblematic in both. Thirdly, there are social justice concerns. Some argue enhancements would only be available to the rich, which are already smarter, thus aggravating social inequality. Neil counter-argues that for such population there would be reduced effects given that cognitive enhancements produce diminishing marginal returns, that many of them are extremely cheap and that, therefore, its effects on inequality would be negligible compared to the effects of traditional means such as education. Fourthly, there is the cheating concern, which Neil contends is not of philosophical interest given that they concern rule violation and not what should be the right rules.

His arguments against the authenticity and inequality worries defuse an objection intended to target cognitive enhancement alone by showing the enhancement is no more problematic than a functionally similar traditional mean, thereby revealing the usefulness of his proposed Parity Principle. Christoph Bublitz and Reinhard Merkel seemed to have argued against this principle by claiming to be an asymmetry, non-traditional means exert changes that bypass the mechanisms that were supposed to process them whereas traditional means do not. Hence, one would not equally control one’s mental states produced by enhancement, and they would not be reflective of who one genuinely is. However, Neil asks us to consider again two isofunctional ways of causing a person to become, e.g., more individualistic. We could either make that person read Any Rand’s complete works, or wait for him to go to sleep and execute a small neurosurgery in his brain in the areas responsible for individualism. In both scenarios, that person has the same level of control. Moreover, enhancements can also result from intentional controls and traditional means may not – e.g., unintentional emotional or moral improvement enlisted by works of art. Finally, traditional means appear to exert already extrinsic control such as the case of marketing or nudges, which seem to be extremely powerful compared to available enhancements.

I believe, however, that if we realistic assume that the decision processes are done throughout the modifications and not just before them, one could argue that since traditional means take more time and effortful control, it allows one to take many small decisions middle way and exert greater control. For instance, by reading Ayn Rand, someone can reflect middle way and decide individualism is wrong while the same cannot be said of an overnight neurosurgery. Moreover, even if the Parity Principle is true, it could only be true of the cases we have traditional means to compare with and could not be used to analyse the more interesting cases of radical enhancements to produce unparalleled effects.

To listen to the podcast of Prof. Levy’s talk please follow this link.

  • Facebook
  • Twitter
  • Reddit

One Response to “Cognitive Enhancement: Defending the Parity Principle”, St Cross Special Ethics Seminar by Neil Levy

  • Hazem says:

    Reading Ayn Rand and becoming more individualistic obviously entails, as you point out, a different process from becoming more individualistic through some hypothetical brain surgery. The Ayn Rand reader may simply come to realise individualism is not their cup of tea after all. But it’s not clear how this possibility is relevant to the parity principle (PP).

    As I understand it, PP would say: whether we use an external prop like a book to change our way of thinking, or whether we directly alter the brain to achieve the same effect, all else being equal, both these alterations are ethically on par. But once you say that the Ayn Rand reader may have more time to change their mind about individualism (i.e. basically decides not to undergo the change in their thinking), then all else is not equal anymore – and PP is no longer relevant. It is only when all else is equal and a person has already undergone some modification that PP kicks in and points out that worries about authenticity apply equally whether that modification was achieved by “traditional” means or technological ones. At least that’s what I took from it.

Recent Comments

Authors

Subscribe Via Email

Name
Email *

Affiliations