Commentary
Cochrane’s Nativism
Raamy Majeed
University of Manchester, UK
Majeed, Raamy. 2024. “Cochrane’s Nativism.”
Journal of Philosophy of Emotion 5, no. 2: 30-35. https://doi.org/10.33497/2024.Winter.4.
Embedded Files
Where scholarship meets technological innovation . . .
Abstract: The aim of this commentary is to draw out a feature of Cochrane’s view not made explicit in his book and to invite him to say a bit more about it. The topic is nativism about emotion: the view that our emotions are systems/mechanisms/programs hardwired into our brains by evolution, and purpose built to generate certain expressive, physiological and behavioural responses. I argue Cochrane’s nativism is on the surface more attractive than standard nativist views of emotion, as it extends beyond the realm of basic emotions to include more complex emotional phenomena, including sentiments. But the main worry is that it does so at the expense of preserving what was plausible about such views in the first place.
Keywords: basic emotions; nativism; modularity; natural kinds; emotional plasticity
THE NATIVIST DILEMMA
The overall aim of this commentary is to draw out a feature of Cochrane’s (2018) view not made explicit in his book and to invite him to say a bit more about it. This topic is nativism about emotion: the view that our emotions are systems/mechanisms/programs hardwired into our brains by evolution, and purposefully built to generate certain expressive, physiological and behavioural responses.
According to Cochrane’s control theory of affect, affective phenomena (e.g., feelings, emotions, sentiments, etc.) are to be understood in terms of what he calls “valent representations”: very roughly, self-regulating mechanisms which are triggered automatically in response to certain stimuli and aim to either increase or decrease the presence of such stimuli. My focus will be on what the Control Theory tells us about emotion. What is striking is that while the account it offers with respect to emotion is no doubt novel, it shares certain similarities with other nativist conceptions of emotion.
[page 30]
First, Cochrane thinks the self-regulatory mechanisms responsible for affective phenomena are “somewhat modular in nature” (83). Modular accounts presuppose a nativist conception of emotion, where modules are typically thought of as systems/mechanisms/programs hardwired to perform certain tasks (Griffiths 1997). The self-regulatory mechanisms responsible for our emotions seem to fit this profile.
Second, he thinks of such mechanisms as being governed by “innate” concerns. That is, the sorts of stimuli that can trigger our emotional responses are innately-specified in aid of meeting certain regulatory ends. This is not to deny that emotions manifest a degree of flexibility. You fear not just dangerous objects in the Environment of Evolutionary Adaptation (EEA), like snakes and spiders, but things like not saving enough for your retirement, your children failing their exams, your partner getting into a car crash, etc.–things which don’t obviously aid self-regulation. But crucially, such flexibility is to be explained in terms of “how our innate sensitivity to a concern can be tuned towards a new object” (83).
Third, like most nativists, he thinks that emotions form a natural kind. In particular, he claims to “recognise the class of emotions as a natural kind (contra Griffiths 1997)” (66). Cochrane doesn’t actually say what natural kinds are, but on Griffiths’s account, natural kind categories are “projectable” categories that can enter into scientific generalisations. In other words, if a category is a natural kind, the correlations observed in some of its samples can be reliably extrapolated to other instances of that category. We can suppose that emotion categories, like fear and grief, are projectable in this sense because there are certain self-regulating mechanisms which have evolved specifically to regulate these emotions. Moreover, it might be that the broader category emotion also forms a projectable category because its instances all share a certain regulatory profile.
Nativism about emotion now has a relatively long and potted history. Its influence is hard to overestimate, but it also has well-known problems. One of the central philosophical problems can be cashed out in terms of a dilemma. Nativists about emotion can either claim that emotions (or a sub-category of them) form a natural kind at the cost of accommodating many of the affective phenomena we typically count as emotion, or they can accommodate a more diverse range of phenomena, at the cost of arguing that emotions are natural kinds. To elaborate, in attempting to argue that there is a subset of emotion categories that form a natural kind, they limit the sorts of phenomena which can be classified as emotion. Typically, they eschew talk of more complex emotional phenomena, e.g., like jealousy, grief, love, and take these to be either handled by other accounts (e.g., social constructionism) or to be eliminated from our emotion ontologies completely. This, the complaint goes, gives us an impoverished view of emotion. (The point works just as well if we replace the notion of “natural kinds” with “modules,” or emotion-specific “affect programs”).
Unlike most nativist accounts, Cochrane’s view, as he presents it, dodges this dilemma. That is, he argues that emotions are natural kinds and does so all the while attempting to accommodate a more diverse range of phenomena than is typical of nativist accounts of emotion: the regulatory mechanisms account for not just basic emotions, like anger and sadness, but more complex emotional phenomena, e.g., hatred and grief, as well. This, ceteris paribus, makes his account much more attractive than other nativist accounts of emotion. But an important question remains unanswered, is Cochrane justified in presenting his view in this way, i.e., as one that evades the dilemma?
On the face of it, there is enough material in Cochrane’s arsenal to evade it. There are layers upon layers of regulatory mechanisms that might be capable of handling basic and complex emotional phenomena alike. However, I am sceptical that he can actually pull this off, mainly because his account is not adequately sensitive to the sorts of concerns that motivate the dilemma in the first place. There are quite a few different issues at play here, all which have come up in debates about emotional nativism. These issues are well-known, so I shall only mention them briefly and focus mainly on how they arise with respect to the Control Theory. I don’t think
[page 31]
these issues, either by themselves or collectively, refute the Control Theory. However, I do think they are important and should be addressed if the view is to be taken seriously.
NATURAL KINDS
As noted, Cochrane doesn’t actually offer a positive account of what natural kinds are. However, we can note that in most, if not all the ways of conceiving of natural kinds, there is more to being a natural kind than simply having certain things in common. This point is nicely brought out by Griffiths.
In What Emotions Really Are: The Problem of Psychological Categories, Griffiths (1997) makes two claims. First, the vernacular category emotion doesn’t form a natural kind because it picks out three very different kinds of things: basic emotions, cognitively complex emotions, and social pretenses, all of which are brought about by very different kinds of causal mechanisms. Second, that being said, basic emotions do form natural kinds because they stem from certain evolved emotion-specific mechanisms. For example, there is a mechanism for fear, a mechanism for sadness, and so on.
The important point for us is that in making the first claim, Griffiths (2002) is at pains to point out that the question of whether emotions form a natural kind is a different question from whether the concept “emotion” can be given a univocal analysis. He thinks the latter is possible, and only takes his project to show that things which fall under the vernacular category emotion don’t form a natural kind for the purposes of scientific inquiry. For instance, he points out that we can think of emotion as a “putative psychological category of motivational states that exhibit passivity” (Griffiths 1997, 246).
One worry I have is that Cochrane doesn’t seem to recognise this distinction. This is best brought out by his discussion of long-term emotions, or what he calls “sentiments.” Note, the term “sentiment” and “emotion” are often used interchangeably, but this isn’t quite what Cochrane has in mind. By “sentiment,” he means “enduring attitudes that we bear towards specific individuals that coordinate a broad pattern of emotions that are sustained and developed over long periods of time” (Cochrane 2018, 175). Thus understood, sentiments are complex, enduring affective phenomena that aren’t emotions themselves but depend on them. For example, your grief for a loved one who has passed away can involve bouts of intense sadness, anger, panic, numbness, etc. (I shall continue to lump sentiments under the broad banner of “complex emotional phenomena”).
This raises an interesting question: what is the essence of a sentiment such as grief? Since a single sentiment may involve various kinds of feelings and cognitive appraisals, Goldie (2012) argues that instead of searching for an essential feature, it is best to think of sentiments as emergent processes involving various stages, all tied together by a narrative arc. Cochrane remains unconvinced: “I doubt Goldie’s claim that there is no single element underlying grief. On the contrary, I think we can say that the process of grief is sustained by the underlying concern for the dead individual” (Cochrane 2018, 174). Fair enough. It looks like we can give a univocal analysis of the concept of grief. But this is a far cry from arguing that grief is a natural kind.
It is worth noting that Cochrane never explicitly claims that sentiments form a natural kind, only that emotions do. However, there are certainly parts of the book which suggest otherwise. For example, he claims that the “emergent process of grief that Goldie described is actually the product of the kind of long-term adaptation I described earlier” (174). One of my worries is that I’ve not seen anything in the book to justify the claim that grief, or any other sentiment, is actually an adaptation.
[page 32]
On the plus side, this gives us a clearer indication of why sentiments like grief might count as natural kinds on the Control Theory. Very roughly, there is an evolved system/mechanism/program that all instances of a sentiment have in common, which coordinate the complex set of emotional responses we associate with that sentiment. Still, I think it’s worth pressing the question. That is, why think this? Even if we buy into the nativist claim that there are affect programs for certain basic emotions, like sadness, anger, and panic, it does not follow that there is a higher-level “executive” program that coordinates these emotions; one we can take to be a grief program. Most nativists don’t make this extra claim, and it remains unclear why it is warranted here.
COMPLEX EMOTIONAL PHENOMENA
This brings us to the second issue: Can we account for complex emotional phenomena using the sort of machinery we use to explain the simpler ones? According to Cochrane (2018, 12), “Each distinctive form of affective state (e.g., emotions, feelings, sentiments) corresponds with a layer of regulative control.” I don’t deny that regulatory mechanisms might be at play when it comes to complex affective phenomena, such as our sentiments. However, it is less clear to me whether we can understand them purely in terms of such regulatory mechanisms, no matter how multi-layered they turn out to be.
Since there is layer upon layer of regulatory control on this view, the precise nature of the dependence relations between sentiments and emotions isn’t entirely clear. As I understand it, sentiments, such as grief, are to be understood in terms of certain emotional dispositions; or as Cochrane specifies, “dispositions to regulate concerns” (162). This is not to deny their difference. One difference between emotion and sentiment is that emotions are single-track dispositions, whereas sentiments involve multi-track dispositions. For example, while sadness involves certain expressive, physiological and behavioural responses, grief can involve the characteristics of several different emotions; it can include those of sadness, but it can also include those of anger and panic.
There are at least two worries with such a view. First, whether even basic emotions, such as sadness, can be identified with an evolved affect program or module is highly controversial. A major problem is that despite initial optimism, the latest neuroscience evidence suggests that there are no hardwired neurological profiles for basic emotions. For instance, while we used to think that there was a specific amygdala-centred circuity for fear, the latest meta-analyses have failed to reveal any fear circuitry that remains constant across studies (Barrett 2013). To be clear, the amygdala is implicated in fear, but it turns out that it is neither necessary nor sufficient for fear. This is part of the much bigger problem, often known as “the problem of variability.” The main issue is that the empirical data available doesn’t support the existence of any biological markers, either in the body or the brain, associated with basic emotions, such as anger, sadness, fear, happiness, etc., (Scarantino 2015). So, whether there are programs or modules for basic emotions, regulatory or otherwise, remains a point of contention. It would be good to know where Cochrane stands on this problem.
The other problem is that even if we can show that there are modules for basic emotions, whether we can explain complex emotional phenomena in terms of such modules remains a matter of debate. That we can explain complex emotions in terms of blends of basic emotions is an old idea, but to my mind not a very plausible one. A significant problem concerns appraisal criteria: such criteria for complex emotions are too sophisticated to be obtained by merely adding together the criteria for various basic emotions (Griffiths 2002). For example, as Jones (2008) points out, jealousy is triggered not merely by the smell of unfamiliar perfume on your lover, or the recognition that they are once again late from work, but when these features are interpreted as signs of infidelity. So, complex emotions don’t seem to be reducible to basic ones. A similar point can also be made with respect to sentiments. While sentiments might involve dispositions to manifest certain (basic) emotional responses, they can’t be reduced to such emotions. Grief, for instance, is triggered not merely at the sight of certain stimuli, but when such stimuli are interpreted as signs of significant loss.
[page 33]
EMOTIONAL PLASTICITY
The final problem is related, i.e., to what extent can the account accommodate the sorts of flexibility manifested in complex emotions and sentiments? Everyone seems to agree that emotions are “plastic”: that they can be shaped and reshaped by us, as well as our socio-cultural environments (Tappolet 2018). Cochrane does discuss such plasticity to an extent under the title “behavioural flexibility,” which he tries to explain in terms of “how our innate sensitivity to a concern can be tuned towards a new object” (83). But it is doubtful that this fully captures all the ways our emotions are plastic.
The first thing to note here is that Cochrane actually grants far less flexibility to emotions than do recent advocates of the Basic Emotion Theory (BET). According to the old (perhaps caricatured) view, our emotions are input and output closed. That is, both the stimuli that trigger our emotions, as well as the behaviours that are generated as part of the emotional responses, are innately-specified and fixed. Cochrane’s account is an improvement insofar as he thinks of emotions as being input-open. However, crucially, unlike advocates of the New Basic Emotion Theory, e.g., Scarantino (2015), he thinks of our emotions as still being output fixed. As we saw earlier, it remains unclear whether our emotions—even the basic ones—correspond to any fixed behavioural profiles. What’s more, here we can see that our emotions manifest a lot more behavioural flexibility than Cochrane seems to allow. Even an emotion like fear can manifest a complex range of behaviours; arguably much too complex to be innately-specified.
The other point to note here is that advocates of the New BET don’t think their account extends to complex emotions (like jealousy), let alone sentiments (like grief and love). To my mind this is sensible. Short-lived emotional responses, which are the sorts of things envisioned as the outcomes of innate modules, don’t seem to be the right sort of things to account for the output side of complex emotional phenomena. Of course, Cochrane could push back: sentiments just are dispositions to have various short-lived emotional responses. However, as we have seen, the issues raised in section three speak against this.
It is entirely possible that I have misunderstood Cochrane’s position. Perhaps complex emotional phenomena are to be accommodated by various higher-level cognitive processes in conjunction with the sorts of regulatory mechanisms that ground our basic emotions. If so, it would be good to have a bit more clarity on this. So far as I can tell, Cochrane thinks it is self-regulatory mechanisms all the way down. I just don’t see how this is possible without describing all, if not most, mental processing as forms of self-regulation.
CONCLUSION
To end, let me offer what I take to be the general lesson. Nativist positions are plausible to the extent that they offer certain concessions. In broad strokes, nativist theories only accommodate some of the phenomena we normally count as emotion. This often leaves philosophers unsatisfied, but it might be a price worth paying if their views are to be considered plausible. Cochrane’s more ambitious strand of nativism appears to evade this pitfall by extending nativism beyond the realm of basic emotions to more complex emotional phenomena, including sentiments. But the main worry is that it does so at the expense of preserving what was plausible about such views in the first place.
[page 34]
References
Barrett, Lisa Feldman. 2013. “Psychological Construction: The Darwinian Approach to the Science of Emotion.” Emotion Review 5 (4): 379–89. https://doi.org/10.1177/1754073913489753.
Cochrane, Tom. 2018. The Emotional Mind: A Control Theory of Affective States. Cambridge, UK: Cambridge University Press.
Goldie, Peter. 2002. The Emotions: A Philosophical Exploration. Oxford, UK: Oxford University Press.
Griffiths, Paul E. 1997. What Emotions Really Are. Chicago, IL and London, UK: Chicago University Press.
_____. 2004. “Is Emotion a Natural Kind?” In Thinking About Feeling: Contemporary Philosophers on Emotions, edited by Robert C. Solomon, 223-249. New York, NY: Oxford University Press.
Jones, Karen. 2008. “Quick and Smart? Modularity and the Pro-Emotion Consensus." In The Modularity of Emotions, edited by Luc Faucher and Christine Tappolet, 3-27. Alberta, CA: University of Calgary Press.
Scarantino, Andrea. 2015. “Basic Emotions, Psychological Construction and the Problem of Variability." In The Psychological Construction of Emotion, edited by James Russell and Lisa Feldman Barrett, 334-376. New York, NY and London, UK: Guilford Press.
Tappolet, Christine. 2016. Emotions, Value, and Agency. Oxford, UK: Oxford University Press.
[page 35]
Raamy Majeed © 2024
Author email: raamy.majeed[at]manchester.ac.uk
Page updated
Google Sites
Report abuse