The Cognitive Impenetrability of Recalcitrant Emotions

Dr. Raamy Majeed —Postdoctoral Research Fellow on the John Templeton Foundation pro­ject, ‘New Directions in the Study of Mind’ in the Faculty of Philosophy, University of Cambridge and By-Fellow, Churchill College, University of Cambridge

Consider the fol­low­ing emo­tion­al epis­odes. You fear Fido, your neighbour’s dog you judge to be harm­less. You are angry with your col­league, even though you know his remark wasn’t really offens­ive. You are jeal­ous of your partner’s friend, des­pite believ­ing that she does­n’t fancy him. D’Arms and Jacobson (2003) call these recal­cit­rant emo­tions: emo­tions that exist “des­pite the agent’s mak­ing a judg­ment that is in ten­sion with it” (pg. 129). The phe­nomen­on of emo­tion­al recal­cit­rance is said to raise a chal­lenge for the­or­ies of emo­tions. Drawing on the work of Greenspan (1981) and Helm (2001), Brady argues that this chal­lenge is “to explain the sense in which recal­cit­rant emo­tions involve ration­al con­flict or ten­sion” (2009: 413).

Whether we require ration­al con­flict to account for emo­tion­al recal­cit­rance is debat­able. Indeed, much of the present con­tro­versy involves spelling out the pre­cise nature of this con­flict. But con­flict, ration­al or oth­er­wise, isn’t the only fea­ture that is per­tin­ent to the phe­nomen­on. What tends to get neg­lected is pre­cisely what gives these emo­tions their name, viz. their recal­cit­rance; their per­sist­ent nature. To elab­or­ate, emo­tion­al epis­odes, by their very nature, are epis­od­ic, and we shouldn’t expect recal­cit­rant emo­tions to last any longer than non-recalcitrant ones. Nevertheless, it is in the very nature of recal­cit­rant emo­tions that they are mul­ish, that they don’t suc­cumb to our judge­ments – i.e. to the extent that these emo­tion­al epis­odes last.

Here is an example. Suppose I judge that fly­ing is safe, but feel instantly afraid as soon as my plane starts to take off. But sup­pose, also, that once I real­ize that my fear is irra­tion­al, or at least, that it is in ten­sion with my judge­ment, my fear dis­sip­ates. This, argu­ably, won’t count as an instance of emo­tion­al recal­cit­rance. By con­trast, say I remain fear­ful des­pite my judge­ment. I keep think­ing to myself, ‘I know this is safe’, and yet I con­tin­ue to feel afraid. This, I ven­ture, bet­ter cap­tures what we mean by emo­tion­al recal­cit­rance. Mutatis mutandis for being afraid of Fido, being jeal­ous of your partner’s friend etc. All famil­i­ar cases of emo­tion­al recal­cit­rance seem to share this per­sist­ent fea­ture. The ques­tion is, what accounts for it?

My hypo­thes­is is this: emo­tions are recal­cit­rant to the extent that they are cog­nit­ively impen­et­rable. According to Goldie, “someone’s emo­tion or emo­tion­al exper­i­ence is cog­nit­ively pen­et­rable only if it can be affected by his rel­ev­ant beliefs” (2000: 76). So far as I can tell, the first to dis­cuss the cog­nit­ive (im)penetrability of emo­tions is Griffiths (1990, 1997), who takes one of the advant­ages of his the­ory to be pre­cisely that it accounts for recal­cit­rant emo­tions, or what he calls ‘irra­tion­al emotions’.

Griffiths’s explan­a­tion of emo­tion­al recal­cit­rance is neg­lected by much of the cur­rent lit­er­at­ure on the phe­nomen­on. This is war­ran­ted in one respect. Griffiths doesn’t account for the sense in which recal­cit­rant emo­tions involve ration­al con­flict, which, as men­tioned earli­er, is one of the cent­ral con­tro­ver­sies. But there is a way in which the neg­lect is unwar­ran­ted. This has to do with the charge that his account makes emo­tions too piecemeal.

To elab­or­ate, one of the most con­tro­ver­sial fea­tures of Griffiths’s account of emo­tions more gen­er­ally is that it div­vies up emo­tions into three broad types, only one of which forms a nat­ur­al kind. These are the set of evolved adapt­ive ‘affect-program’ responses, which are, more or less, cog­nit­ively impen­et­rable. They are sur­prise, fear, anger, dis­gust, sad­ness and joy. The rest are ‘high­er cog­nit­ive emo­tions’, which are cog­nit­ively pen­et­rable, like jeal­ousy, shame etc., or social con­struc­tions that are ‘essen­tially pre­tences’, e.g. romantic love.

This account, argu­ably, does make emo­tions too piece­meal, but to reject the hypo­thes­is that recal­cit­rant emo­tions are cog­nit­ively impen­et­rable for this reas­on is to throw the baby out with the bathwa­ter. Let us be neut­ral as to what emo­tions actu­ally are, as well as to the kinds of emo­tions that can be cog­nit­ively impen­et­rable. I think we can remain thus neut­ral, and still bor­row some of Griffiths’s insights con­cern­ing the cog­nit­ive impen­et­rab­il­ity of recal­cit­rant emo­tions to explain their recalcitrance.

Leaving aside the Ekman-esque notion that there are a set of basic emo­tions from which all oth­er emo­tions arise, we can fol­low Griffiths in sup­pos­ing that emo­tions, indeed the very same kind of emo­tions, can be brought about in dis­tinct ways. Take, for instance, the affect-program responses. The pro­cesses that typ­ic­ally give rise to them, as well as these responses them­selves, are what Griffiths claims is cog­nit­ively impen­et­rable. But he notes that they can also be triggered by pro­cesses that are cog­nit­ively pen­et­rable. In fact, he is clear that the former doesn’t rule out the lat­ter: “[t]he exist­ence of a rel­at­ively unin­tel­li­gent, ded­ic­ated mech­an­ism does not imply that higher-level cog­nit­ive pro­cesses can­not ini­ti­ate the same events” (1990: 187).

Griffiths exploits this account to explain emo­tion­al recal­cit­rance. In brief, the phe­nomen­on occurs when an affect-program response is triggered without the cog­nit­ive pro­cess of belief-fixation that gives rise to judge­ment. For example, “[if] only the affect-program sys­tem classes the stim­u­lus as a danger, the sub­ject will exhib­it the symp­toms of fear, but will deny mak­ing the judge­ments which folk the­ory sup­poses to be impli­cit in the emo­tion” (1990: 191).

This explan­a­tion isn’t sup­posed to provide us with an account of what recal­cit­rant emo­tions are; what picks them out as a type. Rather, for Griffiths, it gives us a ‘the­ory’ of them; we have an explan­a­tion for their occur­rence. Regardless of wheth­er this the­ory is adequate, it is my view that the work such an explan­a­tion can be fur­ther put towards is to explain the recal­cit­rant nature of recal­cit­rant emo­tions. While the affect-program responses don’t always run in tan­dem with the cog­nit­ive pro­cesses involved in belief-fixation, what explains the per­sist­ent nature of these responses is that they, as well as the pro­cesses that give rise to them, are cog­nit­ively impen­et­rable. Moreover, cog­nit­ive pen­et­rab­il­ity admits of degrees. Thus, the extent to which such responses are recal­cit­rant will depend on the extent to which they, as well as the pro­cesses that give rise to them, are cog­nit­ively impenetrable.

One of the advant­ages of his the­ory, accord­ing to Griffiths, is that “[t]he occur­rence of emo­tions in the absence of suit­able beliefs is con­ver­ted from a philo­soph­ers’ para­dox into a prac­tic­al sub­ject for psy­cho­lo­gic­al invest­ig­a­tion” (1990: 192). The present explan­a­tion is sim­il­arly advant­age­ous in that it provides an explan­a­tion of emo­tion­al recal­cit­rance that is empir­ic­ally veri­fi­able. But by the same token, the explan­a­tion is only of interest to the extent that it is empir­ic­ally plaus­ible. The evid­ence is far from con­clus­ive, but there is good reas­on to think we are on the right track.

McRae et al. (2012) sought to test “wheth­er the way an emo­tion is gen­er­ated influ­ences the impact of sub­sequent emo­tion reg­u­lat­ory efforts” (pg. 253). Emotions can be triggered ‘bot­tom up’, i.e. in response to per­cept­ible prop­er­ties of a stim­u­lus, or ‘top down’, i.e. in response to cog­nit­ive apprais­als of an event. They took their find­ings to “sug­gest that top-down gen­er­ated emo­tions are more suc­cess­fully down-regulated by reapprais­al than bottom-up emo­tions” (pg. 259). Emotions gen­er­ated bottom-up, then, appear to behave as if they are cog­nit­ively impen­et­rable; or at least, as if they are less pen­et­rable than ones gen­er­ated top-down. Insofar as any of the emo­tions thus gen­er­ated con­flict (in the rel­ev­ant sense) with an eval­u­at­ive judge­ment, we have an instance of emo­tion­al recal­cit­rance. Run these thoughts togeth­er, and they imply that recal­cit­rant emo­tions are recal­cit­rant to the extent that they are cog­nit­ively impenetrable.



Brady, M. S. (2009). ‘The Irrationality of Recalcitrant Emotions’. Philosophical Studies 145: 413–30.

D’Arms, J., & Jacobson, D. (2003). ‘The Significance of Recalcitrant Emotion’. In A. Hatzimoysis (Ed.), Philosophy and the Emotions. Cambridge: Cambridge University Press.

Goldie, P. (2000). The Emotions: A Philosophical Exploration. Oxford University Press.

Greenspan, P. S. (1981). ‘Emotions as Evaluations’. Pacific Philosophical Quarterly 62: 158–69.

Griffiths, P. E. (1990). ‘Modularity, and the Psychoevolutionary Theory of Emotion’. Biology and Philosophy 5: 175–96.

—- (1997). What Emotions Really Are. Chicago University Press.

Helm, B. (2001). Emotional Reason. Cambridge University Press.

McRae, K., Misra, S., Prasad, A. K., Pereira, S. C., Gross, J. J. (2012). ‘Bottom-up and Top-down Emotion Generation: Implications for Emotion Regulation’. Social Cognitive and Affective Neuroscience 7: 253–62.