Additive Rationality


Additive Theories of Rationality

If, as McDowell maintains, the perceptual capacity of a mature rational subject is actualized, in the basic case, in a kind of state whose presence itself involves the actualization of conceptual capacities, then we cannot explain what it is for a mature rational subject to perceive without making reference to her rational capacities. And if capacities differ in nature inasmuch as their actualizations differ in nature, we cannot explain the nature of the perceptual capacity of a mature rational subject without reference to these latter capacities. In that case, the additive approach cannot succeed, for the additive theorist’s project is to say, not merely what lies in the developmental background of a mature rational subject’s perceptual capacity, but what this capacity consists in (8)

  • What the heck is the argument here? I can’t really follow it or put in into other terms.
  • The argument also seems to presume a form of reductionism (e.g. “the additive theorist’s project is to say…what this [rational] capacity consists in”). But an additive theorist might very well think that one needs a developmental conception of rationality that nevertheless is non-reductive

Problem One - Interaction

For what is needed is not just any sort of constraint on the subject’s judging; what is needed is something intelligible as a constraint from the subject’s own point of view – something she could see as a reason for judging the world to be thus-and-so, if she were to reflect on the question “Why should I believe that?” But to suppose that the content of perceptual experience is nonconceptual is to conceive of it in a way that rules out its playing such a role (11).

  • the first thing to note about this way of motivating M’s project is that it will immediately seem, to many, to beg all sorts of questions about justification.
    • what is it to be “intelligible as a constraint from the subject’s own point of view”?
    • why would the fact that the content of a perceptual experience is non-conceptual rule out it’s playing a justificatory role (at least in principle)?
    • what exactly is it to claim that the content of an experience is non-conceptual?

I would have thought that the main claim of the additive theorist would be that a desire or perception of type T could be had independently of a subject’s possession of some further rational capacity to reflect, be conscious of, judge, or otherwise engage in activity which we consider rational.

But Boyle seems to take the additive theorist’s claim as one conceding whether some token representational state could occur in the absence of any connection or engagement of other rational faculties. But surely the additive theorist doesn’t want to make a claim regarding such things. Her only concern is whether the conditions for identifying some mental state remain stable given the introduction of more sophisticated cognitive capacities. Specifically, she wants to know whether states which occur in beings which lack some set of cognitive capacities (e.g. the rational ones) also occur (with respect to type) in beings which possess those capacities, and perhaps necessarily so, insofar as the more sophisticated capacities presuppose the possession of the less sophisticated ones.

Problem Two - Unity

What I am calling the Unity Problem is a difficulty about how to account for the intuitive idea that the same subject is both a certain animal and the subject who thinks. The animal is supposed to be the subject of various perceptions and desires. The thinker is supposed to be the subject of various reflective thoughts. Is the very same subject the locus of all these activities? That of course is what everyone wants to say. No additive theorist would want to claim that the “I” that thinks stands over an “it” that perceives and desires. But given how the additive theorist conceives of our powers of perception and desire, it is not clear how the unity of this subject is secured. (37)


  • Framing

    Does the whole frame in terms of additive vs. genus/species beg important questions against the naturalist?

  • Justification

    The two objections from interaction and unity seem to depend on the idea that there can be no distinction between the availability of a justifier and the having of a reason in a further, reflectively accessible sense. But there isn’t any argument for this, at least as far as I can tell.

    Pryor 2005 makes the distinction between a justification maker and a justification shower. The notion of a justification-maker is a notion of something that makes a belief just, reasonable, or rationally held (194). The notion of a justification-shower is the notion of something that a subject is in a position to appeal to in the act of proving or showing that some belief they hold is justified, reasonable, true, rational, etc.

    So one way of understanding McDowell and Boyle here is to understand them as arguing that nothing may be a justification-maker unless it is also a justification-shower. Another way of putting this is that nothing can be a reason for a belief unless it is also a reason the subject “has” in the sense that it is something upon which the subject’s belief could be explicitly based.

  • Burgean alternative

    Does Burge present a view that is a plausible intermediary between the additive and transformative views? Partly this hinges on whether Burge’s notion of “entitlement” could be seen as a legitimate form of epistemic warrant on Boyle’s view. Of course, if it can’t (unless there are good non-question begging reasons why), this may just show that the two frameworks aren’t really in a position to talk to one another.

An Analogy

One thing I don’t like about the “additive” moniker is that it conflates different senses of “addition”. For example, consider two liquids, water and oil. The oil sits upon the water and doesn’t mix with it. It doesn’t change it in any way. We can study the properties of each liquid in isolation of the other and explain why they interact as they do.

Now, consider water and salt. The salt is added to the water like the oil. But it doesn’t remain separate. Due the properties of both salt and water, which we can study independently, the salt “integrates” with the water to form a new thing, a salt solution. Why isn’t the set of capacities marking rational from non-rational more like the salt than like the oil? If it is more like the salt then wouldn’t this allow for independent study of the liquid in which the salt is dissolved for insight into the nature of rationality and into its role in this combination of rational and non-rational capacities.

General Discussion

What is the additive/transformative distinction? (p. 5)

Historical roots

The basic conception of “transformative” rationality comes from a particular passage in Aristotle’s Metaphysics:

by genus I mean that one identical thing which is predicated of both and is differentiated in no merely accidental way… For not only must the common nature attach to the different things, e.g. not only must both be animals, but this very animality must also be different for each… For I give the name of ‘difference in the genus’ to an otherness which makes the genus itself other. (Aristotle 1984, X. 8 (1057b39–1058a7): 1671; quoted in boyle2016, 6)

So the specific difference which carves out one species from another in the same genus, also affects the way in which the genus is realized in the species. Thus the nature of animality in rational beings is going to be different from that in non-rational beings.

This immediately raises the question as to whether a “genus” then has any explanatory force. If animality is different in different species, then to what extent can they “share” the same genus animal? This problem is somewhat undersold in Boyle’s description.

The difference between additive and transformative theories of rationality is not that additive theorists admit, whereas transformative theorists deny, that the minds of rational and nonrational creatures have something in common. As McDowell observes, the dispute is about how to understand this commonality. Additive theorists advocate a certain way of understanding what we have in common with nonrational animals: they hold that there must be a distinguishable factor in rational powers of perception and action that is of the very same kind as the factor that wholly constitutes merely animal powers of perception and action. Transformative theorists, by contrast, locate the similarity between rational and nonrational mentality in a different sort of explanatory structure. They hold that rational mentality and nonrational mentality are different species of the genus of animal mentality. What the two ‘have in common’, on this view, is not a separable factor that is present in both, but a generic structure that is realized in different ways in the two cases. Rational and nonrational animals do not share in the sensory and conative powers of nonrational animals; they share in the sensory and conative powers of animals, where this is a generic category of power that admits of two more specific sorts of realization (boyle2016, 5/531)

The distinction

shared “generic structure” realized in different ways
non-generic structure which is (i) “distinguishable” and (ii) shared by members of different species of the same genus

Boyle also puts the additive view in terms of a “two systems view”:

(1) a more primordial system that we share with nonrational creatures (in one case, a perceptual system that adjusts our behavioral dispositions in response to changing sensory inputs; in the other, a motivational system that translates desires for things into behavior directed toward the pursuit of them); and (2) a ‘reasoning system’ that ‘monitors’ the activities of the more primordial system, ‘assesses’ the rational warrant for those activities, and ‘regulates’ these activities in response to its assessments. (boyle2016, 3)

Does the distinction beg any questions?

Is the genus/species distinction question-begging?

Contemporary biologists find species essentialism implausible.

essentialism requires that species-specific traits be shared by all and only members of the species, but it is difficult to find biological traits that satisfy this condition – i.e. species membership is fragile
essentialism stands in tension with evolutionary theory, since species living in similar habitats will face similar evolutionary pressures, and thus possess similar traits – i.e. species features are ubiquitous in a given environment
essences aren’t necessary to explain variation in a population. Instead of explaining variation by appeal to essence plus (environmental) interference, the biologist can simply appeal to gene frequencies in a population plus selection and drift forces – i.e. species essences are superfluous

Since “additive” theorists are presumably going to opt for what our best biological science tells us concerning issues of taxonomy, it is difficult to see how there could even be a non-question-begging means of getting Boyle’s objections off the ground. The whole distinction is set up in a framework that the naturalist is likely to deny.

What are the problems for the additive theorist?

Interaction problem

When I speak of an interaction problem, this is the sort of difficulty I have in mind. Such a problem will arise for any view that posits a situation with the following structure:

(Normal Explanation) For any fact F of type T1, a normal explanation of F must appeal to a fact that relates to F in way W.

(Non-disruptive Influence) Facts of type T2 can normally explain facts of type T1

(System Externality) Facts of type T2 do not relate to facts of type T1 in way W.

A view that is committed to versions of these three theses is committed to an incoherent position. But this, I will argue, is exactly the sort of situation that additive theories of rationality characteristically produce: one in which a certain system is supposed to be non-disruptively influenced by a power whose operations are conceived in such a way that they could only influence the system by disrupting it. (boyle2016, 11)

Boyle then pursues a particular version of this incompatible trio by examining desire-based theories of reasons. He appeals to Quinn’s radio example to motivate the claim that normal human desire involves the presentation of the object of desire as “prima facie meriting the endorsement of rational reflection” (boyle2016, 15). He then puts the objection this way:

A normal human desire presents its object as prima facie to-be-pursued, where this means something like prima facie meriting the endorsement of rational reflection. Reflection may of course overrule immediate desire, but if our account of desire does not make intelligible how our desires can, so to speak, present a verdict that our reason must recognize as at least presumptively significant, then we lose the intuitive contrast between ordinary desires and Quinn-ish impulses to turn on radios. But a view on which my desiring E does not normally involve E’s being presented as desirable could, it seems, only represent desire in this way: as a disposition to pursue E that did itself not engage my sense of what there is reason to do. (boyle2016, 16)

  • Replies to Boyle

    • None of the discussion of the Quinn example seem particularly compelling.

      • Why is the example being discussed the only relevant theory – why would an additive theorist accept the view under discussion in the first place (i.e. wouldn’t they think that desires are reasons?)
    • Even if we accept the argument against the proffered theory, why think this generalizes to all, or even most, additive theories?

Unity problem

There is no straightforward statement of the problem. Here is one attempt:

The question at issue here is not one the eye can judge, it concerns how rightly to conceive of a rational animal, whether it is properly regarded as a single subjectivity of which both desires and choices are predicated, or two subjectivities standing in a relation. The difficulty for the additive theorist is to explain how, if the person’s desiderative capacities are intrinsically independent of his or her capacity for reflective choice, the operations of these two capacities can express one and the same point of view: not an ‘I’ confronted with a resistant ‘it’, but a single, evolving conception of what is to be done. To the extent that this question remains unanswered, the additive approach faces difficulties in explaining what it is for me to be a certain animal analogous to the difficulties Cartesianism faces in explaining what it is for me to be an embodied living thing. (boyle2016, 23)

One straightforward means of responding to the problem is simply to say that there is no guarantee. People are potentially just aggregates of drives – as Nietszche himself argued. Why think we are anything more – i.e. why think that this conception Boyle is starting from is anything other than a philosopher’s fantasy?


In thinking recently about Kant’s view of the basing relation and the way in which his view calls on various conceptions of causation, including formal and final causation, I’m starting to wonder whether I’m not also committed to difference-in-kind/transformative thesis. If what it takes to individuate the faculties involved in, e.g., perception calls on formal and final causes that are lacking in non-rational beings, then the faculties so individuated ipso facto cannot be shared by non-rational beings.

But I don’t think that Kant thinks of the faculties of sensibility as being so individuated, nor that one would need to rationally individuate sensory faculties (i.e. appeal to rational formal and final causes) in order to explain the possibility of intuition or episodic memory or association in non-rational beings. Moreover, we share with non-rational animals the same associative tendencies.

But here’s a way in which I agree with Boyle and Conant: taking two animals, one rational and the other not, it might be the case that they share mental states of various kinds (though this is complicated on the Boyle/Conant picture since it looks like mere animals lack intentional states altogether), but the dispositions by means of which these states are tokened or connected are going to be individuated in altogether different ways, since in the one case they rely on a set of formal and final causes based on a being’s rational capacities and in the other case they do not. It would be worth working out in detail relative to some specific case how this might work.

Note too that this looks all or nothing. But the tricky thing we see in cognitive ethology is that various kinds of rational capacities look to be coming in degree rather than absolutely.


The “transformative” notion is refuted as a reading of Kant from consideration of the following:

  1. If I am right concerning the difference between a receptive and spontaneous faculty then the transformative reading is committed to construing sensibility as a form of receptive spontaneity, which is a contradiction in terms.
  2. Kant’s account of error involves the rational faculty being interfered with by the non-rational; this account makes little (or even less) sense on the transformative reading
  3. Kant’s account of “necessitation” (and related concepts, like the holy will), makes little sense if our sensibility is itself a manifestation of our rational capacity.
  • [convert-to-zettel/objective-representation]
  • [convert-to-zettel/animal-representation]
  • [convert-to-zettel/discursivity-and-rationality]
  • [2019-0310-1502-kant-causation]
  • [2019-0125-1651-rationality]
  • [2019-0130-1325-rational-vs-irrational]

Icon by Nun from The Noun Project. Website built with Org-mode, Hugo, and Netlify.