The central rhetorical device of the target article is the similarity between selfish genes that often work against the interests of the organism that carries them and our selfish goals that often work against our interests. Although this selfish gene/selfish goal analogy is supposed to be merely illustrative (and the authors presumably do not endorse any strong version of evolutionary epistemology), I will argue that the analogy does more harm than good. More precisely, I will argue that Huang & Bargh's (H&B's) definition of goals as “mental representations of desired end-states” (sect. 2.2, para. 2) is ambiguous, and the selfish gene analogy pushes the authors to resolve this ambiguity in the less promising direction.
The definition of goals as “mental representations of desired end-states” is ambiguous between what could be called “specific goals,” where the represented end-state is the end-state of a token action I am about to perform, and “unspecific goals,” where it is the represented end-state of an action-type that would satisfy my desire (without specifying how that would be achieved). To use an example, if I am hungry, I can have an unspecific goal of filling my stomach with something (which is left unspecified) or a specific goal of eating this particular piece of chocolate cake right in front of me. The ambiguity is not just a feature of the definition H&B give, but they also use examples of these two different kinds of goals interchangeably throughout the article.
The specific goal interpretation of the Selfish Goal Theory makes the analogy with selfish genes very straightforward. Goals in this sense are self-contained entities that strive to be fulfilled, often at the expense of our interests. And this sounds very similar to the conception of genes as self-contained entities that replicate themselves. Leaving the huge debate in evolutionary biology and philosophy of biology aside about just how self-contained entities genes are and whether they should really be taken to replicate themselves (but see Godfrey-Smith Reference Godfrey-Smith2000; Nanay Reference Nanay2002; Reference Nanay2011), what really matters for the purposes of the target article is that in only a smallish portion of the examples do H&B use a specific goal in the sense of being self-contained. In the majority of the examples, the goal is very much unspecific and it is specified by mental states that are not part of the goal itself.
The difference boils down to a difference in what general picture of motivation one endorses. If we accept the self-sufficient specific goal picture, then the only mental state that is needed to motivate us to act is this specific goal. If I have a specific goal of eating this particular chocolate cake, this motivates me to act. But here is a more flexible model that H&B often slide into (and they are right to do so). Two things are needed for motivating us to act: an unspecific goal (of, say, wanting to eat something) and a separate representation of something edible in front of us. If this unspecific goal and this “action-oriented representation” are combined, the action is performed.
Just what these “action-oriented representations” are (for example, whether they can be perceptual states) I would like to leave open. They are representations of the objects in the agent's environment that could be used to achieve the unspecific goal. They are not themselves “representations of a desired end-state” – they are representations of means of achieving this desired end-state. And they do not themselves motivate us to act. We are only motivated to act if we have both an unspecific goal and an “action-oriented representation” (see Jeannerod Reference Jeannerod1997, who calls these “visuomotor representations” and Nanay Reference Nanay2013 who calls them “pragmatic representations”).
Which picture of motivation should we choose? H&B themselves seem to be conflicted about this – they seem to go back and forth between these two frameworks, depending on the examples they analyze. But the overall selfish gene analogy pushes them towards the self-sufficient specific goal picture. I argue that this is a mistake and the logic of many of their own examples would demand that they use the more flexible unspecific goal plus action-oriented representation picture.
Everything H&B say can be formulated in this framework: The unspecific goal can be unconscious, and the action-oriented representation can also be (and most often it is) unconscious (Dehaene et al. Reference Dehaene, Naccache, Le Clec'H, Koechlin, Mueller, Dehaene-Lambertz, van de Moortele and Le Bihan1998; Goodale Reference Goodale2011; Jeannerod Reference Jeannerod1997; Nanay Reference Nanay2013). So we get a more nuanced picture about the relation between the conscious and the unconscious processes that lead to action. To put it very simply, both the unspecific goal and the action-oriented representation can be unconscious. And often both of them are.
Sometimes we have an unspecific goal and this influences the action-oriented representation we form (say, we are hungry and look around in the fridge to see what we can eat). Some other time, the action-oriented representation comes first and this triggers the unspecific goal (say, you walk past a café and see a delicious cake in the window that makes you want to eat). H&B themselves analyze those very interactions in the target article, but they can only be made sense of if there are two mental states that can interact in these two different ways (and not just one), that is, in the more flexible unspecific goal plus action-oriented representation framework. They would be better off using this way of framing their claims instead of the more catchy but ultimately misleading (and from an evolutionary biology/philosophy of biology point of view, somewhat suspicious) analogy with selfish genes.
You have
Access
The central rhetorical device of the target article is the similarity between selfish genes that often work against the interests of the organism that carries them and our selfish goals that often work against our interests. Although this selfish gene/selfish goal analogy is supposed to be merely illustrative (and the authors presumably do not endorse any strong version of evolutionary epistemology), I will argue that the analogy does more harm than good. More precisely, I will argue that Huang & Bargh's (H&B's) definition of goals as “mental representations of desired end-states” (sect. 2.2, para. 2) is ambiguous, and the selfish gene analogy pushes the authors to resolve this ambiguity in the less promising direction.
The definition of goals as “mental representations of desired end-states” is ambiguous between what could be called “specific goals,” where the represented end-state is the end-state of a token action I am about to perform, and “unspecific goals,” where it is the represented end-state of an action-type that would satisfy my desire (without specifying how that would be achieved). To use an example, if I am hungry, I can have an unspecific goal of filling my stomach with something (which is left unspecified) or a specific goal of eating this particular piece of chocolate cake right in front of me. The ambiguity is not just a feature of the definition H&B give, but they also use examples of these two different kinds of goals interchangeably throughout the article.
The specific goal interpretation of the Selfish Goal Theory makes the analogy with selfish genes very straightforward. Goals in this sense are self-contained entities that strive to be fulfilled, often at the expense of our interests. And this sounds very similar to the conception of genes as self-contained entities that replicate themselves. Leaving the huge debate in evolutionary biology and philosophy of biology aside about just how self-contained entities genes are and whether they should really be taken to replicate themselves (but see Godfrey-Smith Reference Godfrey-Smith2000; Nanay Reference Nanay2002; Reference Nanay2011), what really matters for the purposes of the target article is that in only a smallish portion of the examples do H&B use a specific goal in the sense of being self-contained. In the majority of the examples, the goal is very much unspecific and it is specified by mental states that are not part of the goal itself.
The difference boils down to a difference in what general picture of motivation one endorses. If we accept the self-sufficient specific goal picture, then the only mental state that is needed to motivate us to act is this specific goal. If I have a specific goal of eating this particular chocolate cake, this motivates me to act. But here is a more flexible model that H&B often slide into (and they are right to do so). Two things are needed for motivating us to act: an unspecific goal (of, say, wanting to eat something) and a separate representation of something edible in front of us. If this unspecific goal and this “action-oriented representation” are combined, the action is performed.
Just what these “action-oriented representations” are (for example, whether they can be perceptual states) I would like to leave open. They are representations of the objects in the agent's environment that could be used to achieve the unspecific goal. They are not themselves “representations of a desired end-state” – they are representations of means of achieving this desired end-state. And they do not themselves motivate us to act. We are only motivated to act if we have both an unspecific goal and an “action-oriented representation” (see Jeannerod Reference Jeannerod1997, who calls these “visuomotor representations” and Nanay Reference Nanay2013 who calls them “pragmatic representations”).
Which picture of motivation should we choose? H&B themselves seem to be conflicted about this – they seem to go back and forth between these two frameworks, depending on the examples they analyze. But the overall selfish gene analogy pushes them towards the self-sufficient specific goal picture. I argue that this is a mistake and the logic of many of their own examples would demand that they use the more flexible unspecific goal plus action-oriented representation picture.
Everything H&B say can be formulated in this framework: The unspecific goal can be unconscious, and the action-oriented representation can also be (and most often it is) unconscious (Dehaene et al. Reference Dehaene, Naccache, Le Clec'H, Koechlin, Mueller, Dehaene-Lambertz, van de Moortele and Le Bihan1998; Goodale Reference Goodale2011; Jeannerod Reference Jeannerod1997; Nanay Reference Nanay2013). So we get a more nuanced picture about the relation between the conscious and the unconscious processes that lead to action. To put it very simply, both the unspecific goal and the action-oriented representation can be unconscious. And often both of them are.
Sometimes we have an unspecific goal and this influences the action-oriented representation we form (say, we are hungry and look around in the fridge to see what we can eat). Some other time, the action-oriented representation comes first and this triggers the unspecific goal (say, you walk past a café and see a delicious cake in the window that makes you want to eat). H&B themselves analyze those very interactions in the target article, but they can only be made sense of if there are two mental states that can interact in these two different ways (and not just one), that is, in the more flexible unspecific goal plus action-oriented representation framework. They would be better off using this way of framing their claims instead of the more catchy but ultimately misleading (and from an evolutionary biology/philosophy of biology point of view, somewhat suspicious) analogy with selfish genes.