Overview
The general theme of my work is understanding who we are, and how our brains could--and should--inform this understanding. I approach these topics from a naturalist and and anti-intellectualist perspective. That is, I believe that we are physical beings and (smart) animals, whose minds can be understood through the study of psychology and neuroscience. And I believe that the aspects of our minds most important to who we are are those we share with animals--feeling or affect, motivation, and attention--rather than the putatively 'rational' aspects that we are assumed not to share.
Most of my work falls into two projects: one on the metaphysics of mind, which focuses on reductive physicalism and functionalism, though I have recently branched out a bit into 4E cognition as well. The second is on the nature of affective states (emotion, desire) and psychopathology (addiction). I am also becoming interested in a cluster of issues about the self in moral psychology and action theory, including moral responsibility.
To see the abstract of any of the papers below, just click on the title.
Publications
"What Emotions Really Are (in the Theory of Constructed Emotion" Philosophy of Science 85 (2018):640-59.
Abstract: Recently, Lisa Feldman Barrett and colleagues have introduced the Theory of Constructed Emotions (TCE), in which emotions are constituted by a process of categorizing the self as being in an emotional state. The view, however, has several counterintuitive implications: for instance, a person can have multiple distinct emotions at once. Further, the TCE concludes that emotions are constitutively social phenomena. In this article, I explicate the TCE*, which, while substantially similar to the TCE, makes several distinct claims aimed at avoiding the counterintuitive implications plaguing the TCE. Further, because of the changes that comprise the TCE*, emotions are not constitutively social phenomena.
Links
Links
"Addiction is Not a Natural Kind." Frontiers in Psychiatry 4(2013):123.
Abstract: I argue that addiction is not an appropriate category to support generalizations for the purposes of scientific prediction. That is, addiction is not a natural kind. I discuss the Homeostatic Property Cluster (HPC) theory of kinds, according to which members of a kind share a cluster of properties generated by a common mechanism or set of mechanisms. Leading accounts of addiction in literature fail to offer a mechanism that explains addiction across substances. I discuss popular variants of the disease conception and demonstrate that at least one class of substances that fails to confirm a major prediction of each account. When no mechanism can be found to explain the occurrence of the relevant properties in members of a category, the HPC view suggests that we revise our categories. I discuss options offered by the HPC view, including category revision and category replacement. I then conclude that talk of addiction as a prediction-supporting category should be replaced with categories such as “S-addiction” and “T-addiction,” where S and T are substances or sets of substances of abuse, as these categories are genuine natural kinds.
Links:
Links:
- "Is Addiction a Heterogeneous Condition? Reflections on Pickard's 'The Puropose in Chronic Addiction." (Reply Paper). AJOB Neuroscience 2(2012):52-4.
Abstract: I argue that both the extant neurocognitive theories of addiction, as well as Pickard's purporsive view, fail to capture the phenomenon for at least one substance of abuse. However, Pickard's view is in better shape than its neurocognitive competitors: since Pickard defines addiction in terms of five person-level motivational factors, her view is adaptable into a cluster concept view. This aspect allows for greater flexibility in its predictions/explanations than views which posit and define addiction in terms of a core underlying mechanism.
Links:
Links:
Papers under review/in preparation (email me for the latest draft)
A paper on reward and the NEURoscience of desire (under review)
I propose a novel theory of desire that identifies desires with states of the reward system. The
reward system calculates i) predicted reward in the immediate future and ii) received reward at
the moment (Sutton and Barto 1998). The most fully articulated reward theory of desire to date is
Nomy Arpaly and Tim Schroeder’s (Arpaly and Schroeder 2014; Schroeder 2004). Arpaly and
Schroeder identify intrinsic desires with standing states of the received reward stream, states
which respond to objects that the received reward stream treats as rewarding. I suggest, to reflect
the forward-looking and action-initiating nature of desire, to identify intrinsic desires with
standing states of the predicted reward stream that respond to objects and predict they will be
rewarding. My view intuitively tracks better how we conceive of potentially rewarding objects
before we have obtained them. Further, it identifies desires with the states of the reward system
that activate before the objects of our desires are obtained. If, as I argue, standing desires
generate occurrent ones by activating, like a light bulb turning on, then standing desires must be
a state that activates prior to obtaining the objects of our desires. Otherwise, it would be
impossible for occurrent desires to motivate our actions in the moment.
reward system calculates i) predicted reward in the immediate future and ii) received reward at
the moment (Sutton and Barto 1998). The most fully articulated reward theory of desire to date is
Nomy Arpaly and Tim Schroeder’s (Arpaly and Schroeder 2014; Schroeder 2004). Arpaly and
Schroeder identify intrinsic desires with standing states of the received reward stream, states
which respond to objects that the received reward stream treats as rewarding. I suggest, to reflect
the forward-looking and action-initiating nature of desire, to identify intrinsic desires with
standing states of the predicted reward stream that respond to objects and predict they will be
rewarding. My view intuitively tracks better how we conceive of potentially rewarding objects
before we have obtained them. Further, it identifies desires with the states of the reward system
that activate before the objects of our desires are obtained. If, as I argue, standing desires
generate occurrent ones by activating, like a light bulb turning on, then standing desires must be
a state that activates prior to obtaining the objects of our desires. Otherwise, it would be
impossible for occurrent desires to motivate our actions in the moment.
A Paper on Reduction and species-general mental kinds (under review)
Jaegwon Kim (1998) offers an attractive version of reductive physicalism, but with a catch. Specifically, Kim's view entails giving up on reducing species-general mental kinds (belief, desire, etc.) in favor of, roughly, species-specific mental kinds (human belief, dog belief, Martian belief, etc.). Yet it seems like an empirical question whether a given species-general mental kind is reducible, and thus a reductive theory needs to explain how it is in principle possible for such a kind to be reduced.I propose that Kim's theory can do so by allowing for a broader conception of what counts as a valid physical kind. Kim's ontology, which he borrows from Lewis (1983) is restrictive: specific metals, like aluminum, copper, and gold count as bona fide (i.e., natural) physical kinds but the kind 'metals' does not. I argue, first, that, if species-specific mental kinds can be reduced to physical kinds available in Kim's ontology, then it is at least possible for a species-general kind to reduce to a physical kind like metals. Second, I argue that Kim can adopt a more liberal conception of natural physical kinds--Boyd's (1991) Homeostatic Property Cluster (HPC) Theory of natural kinds--and keep the rest of his reductive view intact. Specifically, I argue i) that these broader kinds like 'metals' count as natural kinds on the HPC conception, and ii) their inclusion in an ontology can be made compatible with Kim's Causal Exclusion principle.
A paper on Functionalism (under review)
Recently, naturalistic functionalists, notably Jake Quilty-Dunn and Eric Mandelbaum have clashed with contemporary dispositionalists, notably Eric Schwitzgebel, about the nature of belief. Unlike previous debates between these two camps, the more contemporary dispositionalism allows for the existence of internal states, so the debate cannot proceed along the same contours as its predecessor between (among others) Ryle and Putnam. I analyze this clash and propose a solution.
I argue that the clash is best understood in terms of an argument about the functional level of description at which belief should be defined. Schwitzgebel is arguing for the primacy of a 'superficial' level at which functional roles are defined in terms of perceptions, behaviors, and the set of (other) mental states minimally necessary to account for all behaviors. Whereas Quilty-Dunn and Mandelbaum are arguing for a 'mechanistic' level at which the superficial-level states are decomposed into functionally delineated components.
I propose that neither can stand on its own, and I articulate a hybrid functionalism that pairs the two together to reap the benefits of each. On my view, functional mental states like beliefs *are* states at the mechanistic level, but i) they are the states that are the 'core realizers' of superficial level functional roles, and ii) those mechanistic level states are belief *in virtue of* being the core realizers of those superficial functional roles.
I argue that the clash is best understood in terms of an argument about the functional level of description at which belief should be defined. Schwitzgebel is arguing for the primacy of a 'superficial' level at which functional roles are defined in terms of perceptions, behaviors, and the set of (other) mental states minimally necessary to account for all behaviors. Whereas Quilty-Dunn and Mandelbaum are arguing for a 'mechanistic' level at which the superficial-level states are decomposed into functionally delineated components.
I propose that neither can stand on its own, and I articulate a hybrid functionalism that pairs the two together to reap the benefits of each. On my view, functional mental states like beliefs *are* states at the mechanistic level, but i) they are the states that are the 'core realizers' of superficial level functional roles, and ii) those mechanistic level states are belief *in virtue of* being the core realizers of those superficial functional roles.
"The Guise of the rewarding: Combining Reward and Guise of the Good Theories of desire"
I argue that a version of the representational ('guise of the good') theory of desire can be combined with a version of the Reward Theory of Desire in a way that can solve issues raised against each view independently.
On the one hand, the guise of the good view argues that desires are perception-like 'seemings' (Tenenbaum 2007; Oddie 2005) or affective states (Railton 2012)--which have the same doxastic status as 'seemings' (Carruthers 2018)--wherein the object desired is evaluated positively in some sense. It has been argued against them that the states posited are necessarily occurrent states, but desires come in standing and occurrent varieties. I argue that what is needed is the identification of a (neuro)cognitive system that can act as the standing state counterpart to quasi-perceptual or affective occurrent states. It must i) stores these subdoxastic evaluations, and ii) generates the states identified with occurrent desires. I propose that the reward system plays just a role.
The Reward Theory of Desire, as articulated by Arpaly and Schroeder (2014), denies that the reward system has any representational content of its own (i.e. no representational 'mode' that it applies to the desired objects that activate it). I argue that such a view of the reward system leaves it open to the same sort of "radioman" objection (Quinn 1993) that philosophers have raised against motivational theories of desire. Understanding the reward system as evaluating stimuli solves this problem.
On the one hand, the guise of the good view argues that desires are perception-like 'seemings' (Tenenbaum 2007; Oddie 2005) or affective states (Railton 2012)--which have the same doxastic status as 'seemings' (Carruthers 2018)--wherein the object desired is evaluated positively in some sense. It has been argued against them that the states posited are necessarily occurrent states, but desires come in standing and occurrent varieties. I argue that what is needed is the identification of a (neuro)cognitive system that can act as the standing state counterpart to quasi-perceptual or affective occurrent states. It must i) stores these subdoxastic evaluations, and ii) generates the states identified with occurrent desires. I propose that the reward system plays just a role.
The Reward Theory of Desire, as articulated by Arpaly and Schroeder (2014), denies that the reward system has any representational content of its own (i.e. no representational 'mode' that it applies to the desired objects that activate it). I argue that such a view of the reward system leaves it open to the same sort of "radioman" objection (Quinn 1993) that philosophers have raised against motivational theories of desire. Understanding the reward system as evaluating stimuli solves this problem.
Papers "Under Construction" i.e., currently being revised in light of feedback from colleagues or a talk. Email me if you'd like a draft, and I will give you the next one I produce. Just know that it might take a bit!
why reduction doesn't matter
Many philosophical issues are thought to turn on the truth of reductive physicalism about the mental. These range from the methodological (how much relevance does neuroscience have for philosophical psychology?) to the 'deep' (is the mental autonomous from the physical?). I argue that these issues do not in fact turn on reduction but rather on on the related notion of realizer independence.
There are two versions of nonreductive physicalism that are grounded in claims of multiple realizability: those that see mental kinds as being realized or realizable in more than one physical kind, and those that see it as being realized in an indefinite or infinite number of physical kinds. I argue that the former version i) is only notationally distinct from a reductive physicalism, and thus ii) carries all of the upshots of a reductive physicalism. The reason i) is true is that, if functional kinds are sufficiently *robust* then their realizers share at least some common physical properties. And whether to call a set of realizers that share a subset of properties a single or multiple kinds is by and large a notational issue.
There are two versions of nonreductive physicalism that are grounded in claims of multiple realizability: those that see mental kinds as being realized or realizable in more than one physical kind, and those that see it as being realized in an indefinite or infinite number of physical kinds. I argue that the former version i) is only notationally distinct from a reductive physicalism, and thus ii) carries all of the upshots of a reductive physicalism. The reason i) is true is that, if functional kinds are sufficiently *robust* then their realizers share at least some common physical properties. And whether to call a set of realizers that share a subset of properties a single or multiple kinds is by and large a notational issue.
extended cognition and conceptual constitution
I propose a novel framework for how mind/brains incorporate external objects that caputures much of the intuition(s) motivating extended cognition while avoiding the claim that the mind literally extends beyond the skin.
In the literature on extended cognition, we are presented with two metaphysical accounts that accompany a mind's being tightly coupled with an external object: a constitutive picture (e.g., Clark and Chalmers 1998) where the object is part of the mind, and a causal picture (e.g., Adams and Aizawa 2001) where the object simply casually interacts with the mind. At least some of the motivation for embracing the former picture is a rejection of the latter.
I offer a third option. Consider the constitutive claim. The idea that if an object partly constitutes some phenomenon then it is a literal part of that phenomenon is certainly one kind of constitution, which I call 'mereological constitution.' But there are other constitutive relations. Consider the property of being an uncle. I am an uncle: the boundary of the uncle is my body. But what makes me an uncle extends beyond my body, out to my brother and his son. They are part of the constitutive basis of my being an uncle, but not parts of the uncle. I call this 'conceptual constitution.' Conceptual constitution is stronger than either casual dependence or content externalism because the external constitutive factor must endure/purdure for the phenomenon in question to do so.
I explore the implications for extended cognition, and argue that in many common examples used to motivate the view, it is plausible that the external objects in question are conceptual, rather than mereological, constituents.
In the literature on extended cognition, we are presented with two metaphysical accounts that accompany a mind's being tightly coupled with an external object: a constitutive picture (e.g., Clark and Chalmers 1998) where the object is part of the mind, and a causal picture (e.g., Adams and Aizawa 2001) where the object simply casually interacts with the mind. At least some of the motivation for embracing the former picture is a rejection of the latter.
I offer a third option. Consider the constitutive claim. The idea that if an object partly constitutes some phenomenon then it is a literal part of that phenomenon is certainly one kind of constitution, which I call 'mereological constitution.' But there are other constitutive relations. Consider the property of being an uncle. I am an uncle: the boundary of the uncle is my body. But what makes me an uncle extends beyond my body, out to my brother and his son. They are part of the constitutive basis of my being an uncle, but not parts of the uncle. I call this 'conceptual constitution.' Conceptual constitution is stronger than either casual dependence or content externalism because the external constitutive factor must endure/purdure for the phenomenon in question to do so.
I explore the implications for extended cognition, and argue that in many common examples used to motivate the view, it is plausible that the external objects in question are conceptual, rather than mereological, constituents.
degrees of senses of responsibility
We intuitively think that moral responsibility--and praise and blame--come in degrees. Authors writing on this topic such as Nelkin (2016) and Coates and Swenson (2013) generally agree that the more *difficult* an action is, then the less responsible (qua blameworthy) we are for it.
Coates and Swenson offer a systematic account of degrees of responsibility based on Fischer and Ravizza's (1998) notion of reasons-responsiveness. They argue that the degree someone's mental mechanism or process is reasons-responsive can be captured by how 'close' (similar) or 'far' (dissimilar) the nearest possible world in which the person acts on that reason is to our actual world.
Nelkin criticizes this account, showing that distance of the closest possible world in which someone acts on a reason cannot distinguish between the difficulty a depressed person has in acting on a reason versus someone with a simple but strong desire not to act on it.
I argue that, nonetheless, Coates and Swenson's account has merit, albeit with a more limited scope. I argue that once we distinguish between *senses* of responsibility (Watson 1989; Fischer and Tognazzini 2011) we can see that Coates and Swenson's picture applies within at least some senses. In particular, once we hold fixed what Fischer and Tognazzini call 'reactive appraisability' which is the extent to which someone is in general an apt target for the reactive attitudes, the degree of reasons-responsiveness tracks what Watson calls 'accountability,' which is the extent to which someone is praiseworthy or blameworthy for a specific action.
I argue that Coates and Swenson’s account makes sense of mitigating blame due to difficulty only when the relevant sense is reactive appraisability: when the difficulty is due to a psychological condition, the degree of reasons-responsiveness can tell us how apt a target the agent is for blame generally.
Coates and Swenson offer a systematic account of degrees of responsibility based on Fischer and Ravizza's (1998) notion of reasons-responsiveness. They argue that the degree someone's mental mechanism or process is reasons-responsive can be captured by how 'close' (similar) or 'far' (dissimilar) the nearest possible world in which the person acts on that reason is to our actual world.
Nelkin criticizes this account, showing that distance of the closest possible world in which someone acts on a reason cannot distinguish between the difficulty a depressed person has in acting on a reason versus someone with a simple but strong desire not to act on it.
I argue that, nonetheless, Coates and Swenson's account has merit, albeit with a more limited scope. I argue that once we distinguish between *senses* of responsibility (Watson 1989; Fischer and Tognazzini 2011) we can see that Coates and Swenson's picture applies within at least some senses. In particular, once we hold fixed what Fischer and Tognazzini call 'reactive appraisability' which is the extent to which someone is in general an apt target for the reactive attitudes, the degree of reasons-responsiveness tracks what Watson calls 'accountability,' which is the extent to which someone is praiseworthy or blameworthy for a specific action.
I argue that Coates and Swenson’s account makes sense of mitigating blame due to difficulty only when the relevant sense is reactive appraisability: when the difficulty is due to a psychological condition, the degree of reasons-responsiveness can tell us how apt a target the agent is for blame generally.