Aliefs are explanatorily valuable

In her paper “Alief and Belief” from 2007 Tamar Gendler has argued for a distinction between belief and what she calls “alief.” She holds that this distinction is indispensable for explaining certain phenomena which involve “belief-behavior-mismatch” (Gendler 2007:5).

In this essay I will argue for the usefulness of the new category of aliefs. First, I point out that the familiar cognitive attitudes of belief and imagination can’t explain the behavior of a subject in belief-behavior mismatch cases. Second, I will show how Tamar Gendler argues that a new concept of cognitive states is necessary and third, how the concept of aliefs solves the dilemma of belief-behavior mismatch. Gendler presents different cases in which aliefs may play a role. I will characterize the two groups of cases with the help of everyday life examples, to show that aliefs are a necessary concept with an explanatory value for belief-behavior mismatch cases. 

In many cases a person consciously believes a certain proposition to be true, but nonetheless acts as if she actually believes it is false. A visitor to the zoo hesitates to walk past a big fish tank with sharks in it, even though she  believes it to be perfectly safe. A person reaches for her purse in her bag, although she knows she left it at home. These cases all differ in some aspects and can be categorized into two groups: 1) Cases in which appearances behaviorally override evidence, and 2) Cases in which habit behaviorally overrides memory. In both cases a subject S judges a proposition P to be true, but in some way fails to act in accordance with P. This can be called a belief-behavior mismatch. It looks as if she considers the proposition to be true, but simultaneously acts as if she represents it as not true. But following the standard assumption of rationality it is not possible to hold both contradictory beliefs at the same time. If it isn’t the subject’s belief that explains her behavior, then what does?

Tamar Gendler argues for a new category of cognitive states, which can explain such discordancy cases: the aliefs. Besides the conscious belief there is an alief, the content of which is of a different kind. It is representational and action guiding, but different from beliefs; It is possible to at the same time believe P to be true and to have an alief the content of which represents it as otherwise. 

I will present one critique of the concept of aliefs, which is given by Eric Mandelbaum. He focuses on one aspect in which Gendler isn’t sure about her notion of aliefs. In one footnote she says that it is not clear as of yet, if aliefs  have a propositional content or not. Mandelbaum points to a dilemma which arises because of this.

Familiar cognitive attitudes belief and imagination 

If a subject S believes a certain proposition P, she believes that it is true that P, and that her belief is non-defective only if, as a matter of fact, it is true that P. Which means that if P turns out to be false, S’ belief of P would no longer hold[2].

Belief: 

S believes that P iff she believes that it is true that P 

Her belief is non-defective only if, as a matter of fact, it is true that P. 

If a subject S imagines that P, she imagines that it is true that P, but the actual truth or falsity of P is explicitly not important for S to successfully imagine it to be. 

Imagine: 

S imagines that p, iff she imagines that it is true, that P. 

The actual truth or falsity of P is not important for S to successfully imagine it to be. 

Both states are similar in one way and different in another. The similarity is that the subject in   both cases is willing to accept the proposition (Gendler 2007, p.13). “Accepting the proposition” means that when she believes or when she imagines P, she regards it as true. But whereas a subject can imagine any proposition to be true, she can only believe what she really takes to be true. This means that imagination is reality insensitive, because it is independent from what we really hold and believe is reality-sensitive, because it depends on what we really hold. 

Let me now introduce two cases of belief-behavior mismatch, which can be categorized into two groups: 1) Those in which appearance overrides evidence, 2) Those in which habit behaviorally overrides memory. By trying to explain these cases with the familiar cognitive attitudes of belief and imagination, we will see that they are not explanatorily appropriate. 

1. Appearances behaviorally overrides evidence 

Imagine a situation where someone is visiting the zoo. The next station is a big fish tank with white sharks in it. The person hesitates to move forward as she spots the dangerous fish in the tank. The guide who is walking her through the zoo assures her that everything is perfectly safe and that she doesn’t have to worry about the glass suddenly breaking. They see lots of other visitors passing the fish tank, amazed by the appearance of the fish and walking by without getting attacked. The person now has evidence that she can enjoy the view of the white sharks, without worrying that the fish tank might break. Given our definition of belief we can correctly ascribe to her the belief that the fish tank is safe. She has quite good evidence this is true and she declares it as true. But still she hesitates, although she claims that she believes to be sure that the glass is safe. Her belief is dissonant with her behavior. The fact that she doesn’t want to step in front of the fish tank seems to reveal that she represents it as not safe.

This representation is not reality-sensitive. The person’s behavior warrants ascribing a different belief, one that is inconsistent with the evidence-based belief. It is not possible to say that she believes it to be both safe and unsafe. There is a need to characterize her representation of the fish tank to be unsafe as something else than belief.

Tamar Gendler attempts to show how it would look like if this behavior were  explained by imagination, in order to show that aliefs can explain it more easily: Maybe the subject is only imagining the fish tank to be unsafe. Perhaps she is only pretending as if the fish tank is unsafe. She imagines it to be unsafe and from that she imagined it to be unsafe, her hesitancy results in. (Gendler 2007:16). This seems quite complicated and unnatural. Her behavior reveals that it is otherwise: Her hesitancy demonstrates that she is really concerned about the fish tank breaking, not only imagining it to break. 

Cases that fall under this group are all similar in one aspect: The subject holds some belief that is evidentially grounded, but this belief is at rational odds with what she does. 

It has been stated hat the subject doesn’t believe the fish tank to be unsafe and doesn’t merely imagine it to be unsafe, we need another concept of representation which explains her behavior. 

Let us now examine the second group of belief-behavior mismatch. One in which habit behaviorally overrides memory. 

2. Habit behaviorally overrides memory 

Suppose subject S accidentally left her purse at home. She noticed its absence while standing in the queue to get into the museum. At the counter she managed to borrow some money from her friend. While telling him how thankful she is for him lending her money, she reaches for her bag, intending to put the money into her purse. Given our definition of belief 

she surely believes that she had left her purse at home. The reason why she borrowed money was because she knew she left it at home. Again, her behavior is dissonant with what she believes. Acting on habit the subject reaches for her purse to put the money into it. Given that we cannot ascribe to her the belief that she left her purse at home and at the same time the belief that she has it in her bag, the representation of her purse being in the bag needs to be characterized as something else. 

Both cases have in common that the subject has at least one belief that would explain a certain behavior, which is incompatible with what the person actually does. Furthermore, it can be said that habit explains why the person does what is at odds with her belief. If we want to characterize the representation which corresponds with this habit and which generates the action, we cannot use belief for it. In this second group of belief-behavior mismatch we also need another concept of representation, which explains her behavior. 

Gendler argues that we can solve this explanation trouble by introducing a new category of mental states. Something that lies between believing and imagining, and is characterized by the following aspects. 

A characterization of aliefs 

An alief is a mental state, which content is a mixture of three elements. It consists of an association between a representation, an affect, and a behavior.[3] In short R-A-B. It occurs in the brain of a subject because of her experiences or her genetic endowments. This means that the subject has undergone some experiences in her life or a certain kind of innateness, which she associates with a R-A-B content. The three sorts of components are activated consciously or non-consciously. When the subject is exposed to some perception or a non-perceptual thought which stimulates a representation of some object, concept, situation or circumstances, it is combined with an experience of some affective or emotional state and some motor routine. It can be said that the subject alieves something. This can happen consciously or non-consciously, so that it might be that the subject isn’t even aware of being in such a state. Whereas beliefs are supposed to represent how things are, and to guide action in accordance with desire, aliefs directly activates behavioral response. 

The definition of an alief is as follows: 

Alief 

An alief is a mental state with associatively-linked content that is representational, affective and behavioral, and that is activated – consciously or non-consciously – by features of the subject’s internal or ambient environment. 

Now let me show you how Gendler explains the examples mentioned above with the help of aliefs. 

1. Appearance behaviorally overrides evidence – explained with aliefs 

Remember in the first case in which subject S consciously believes the fish tank to be perfectly safe. But nevertheless she hesitates, which tells us that she represents it as unsafe. Whereas belief and imagination can’t explain her behavior, the concept of aliefs gives an explanation: Although she believes that the fish tank is safe, she also alieves something different. The subject has undergone some experiences with white sharks. She might have seen how they killed people in movies or heard that they attacked people. The perception of the fish tank with white sharks leads to an association with the undergone experience. It activates the alief, so she represents the fish tank as unsafe (R), feeling the emotion of fear (A) and tries to stay away from it (B). The alief has roughly the following content: “Really dangerous animal. Not safe to walk past. Stay away!” This is fully compatible with believing the fish tank is safe.. The appearance of dangerous sharks overrides the evidence that it is safe and leads to the activation of an alief which represents the fish tank as unsafe. 

2. Habit behaviorally overrides memory – explained with aliefs 

Besides the previous case, where appearance overrides evidence, there are also cases in which habit behaviorally overrides memory.  In this case the subject believes that her purse is not in her bag, but her reaching for it is dissonant with what she believes. If we want to characterize the way she represents the contents of the bag, Gendler says that it is an alief that represents the bag as containing her purse. Although she believed that her purse is at home, she simultaneously alieved something different. She acted on habit, when she reached for the purse instead of putting it somewhere else. She doesn’t believe that the purse is in there, because she has good evidence she forgot it at home. The subject anyhow wants to put the money in her purse, because she automatically associates her bag with her purse, and this association guides her action. The alief has the following content: “Bunch of money. (R) Needs to go into a safe place. (A) Activate purse-retrieval motor routine now. (B)” (Gendler 2007:4). So the characterization of this representation, which corresponds with the habit and which generates the action, is an alief. 

Going through these cases reveals that the introduction of the new concept of aliefs gives a high explanatory value for belief-behavior mismatch, which otherwise can hardly be explained by familiar attitudes such as belief or imagination. 

Critique of the concept of aliefs 

There are several critiques to the new notion of “aliefs”. Given that Gendler leaves open if aliefs have a propositional content or not[1], Eric Mandelbaum proposes the following dilemma in his paper “Against aliefs” (Mandelbaum 2013). On the one hand, if aliefs have propositional content, then it is unclear how they differ from psychological concepts we already have. On the other hand, if aliefs contain no propositional content, he claims, they cannot do the explanatory work that Gendler claims they can.

In some cases it seems that people do use an alief as a proposition in an argument, and infer some other propositions of an alief. For example, imagine a person who will pay a lot of money for a t-shirt which was worn by his favorite celebrity, George Clooney. As Mandelbaum expresses it, the person thinks that the t-shirt contains some “Clooneyish material” (Mandelbaum 2013:18). In such cases the person doesn’t believe that the t-shirt really contains some of George Clooney. The subject is aware that it cannot be the case that a part of George Clooney is in this t-shirt. Rather, she unconsciously thinks that this is the case.. But if she were to be told that this t-shirt had been in the laundry after George Clooney wore it, she wouldn’t be willing to pay that much money for it anymore. She somehow reasons that washing the t-shirt will erase the Clooneyness from it. And here the problem arises: If the alief that the t-shirt containing some of George Clooney were only an associative one, it couldn’t explain such inferences. Mandelbaum says that it is not the case that the subject has strong negative associations with hygiene which could swamp the positive association with Clooneyness (Mandelbaum 2013:17). Rather, it is a propositional state that expresses that the t-shirt contains Clooneyish material and which would allow the inference that it could be washed away.

This example shows that it is a propositional state in which the subject is in, because it acts as a premise in an inference. To assign a belief in this case is not possible, because the subject doesn’t claim she believes the t-shirt really contains Clooneyness. If it is an alief that explains it, it must be a propositional one.

The challenge, according to Mandelbaum, is now to show how aliefs differ from other cognitive states we already use. In order to find out the answer, he introduces a test first presented Fodor. Fodor says that every new class which is built by other properties has to have  its own interesting properties over and above the properties which make it up in order to be established as an own category (Fodor 1983). It needs to do some explanatory work on  its own, which cannot be done by its parts.

As Mandelbaum puts it, in every Gendler case it seems that the explanatory work of the alief is only made by one of the contents of the triple pair of it. In the fish tank case the subject feels afraid even though she knows the tank is safe, the fear that she feels is underwritten by the affect to not walk past it, by the motor response to hesitate. The task now is to find some behavior that is not merely explained by any component parts of the alief, so that some behavior could be explained in terms of aliefs as such.

One of the aspects which might be distinctive only to aliefs and which Gendler claims is what makes them different  from beliefs, is their functional role. The functional role of aliefs is that their content does not seem to be affected by incoming evidence (Mandelbaum 2013:21). In contrast to the functional role of beliefs, Gendler says, which we will drop upon hearing disconfirming evidence. Here Mandelbaum brings up the term of cognitive dissonance, which is the tendency of people to hold on to their beliefs, even when the incoming evidence tells them otherwise. It even seems that in many cases in which a belief is important to the subject, she will, by incoming disconfirming evidence increases her credence in her belief. Examples of this include cult leaders who believe that the end of the world is near, and who somehow increase their credence in their beliefs after their predicted date for the end of the world has  already come and gone.  

Whereas a necessary feature for ‘belief’ is in Gendler view that it is responsive to evidence, Mandelbaum tries to deny this feature by introducing the phenomenon cognitive dissonance.

I think that saying the cognitive dissonance is telling us something about the nature of belief is a category mistake. It is true that humans often hold on to their belief even when incoming evidence should lead them to revise their belief. But this only tells us about what the subject takes to be evidentially relevant and not that belief is in nature not evidence sensitive. Cognitive dissonance therefore tells us something up the rationality of human beings.

Mandelbaum’s criticism is valuable, but all things considered, I think that, based on my discussion, we have sufficient reason to be justified to accept aliefs as a further ontological category. He is mentioning a very important feature about the nature of aliefs, in which even Gendler is not quite sure about. But this criticism is not as harmful for the usefulness of the notion of alief as Mandelbaum states it.

Given that Mandelbaum is not presenting a solution for cases in which subjects aren’t acting accordingly to their beliefs, he is not able to explain these cases. There still some explanatory gap he is facing, which Gendlers concepts of aliefs are doing well. I would not deny that people hold on to their beliefs by incoming disconfirming evidence, but to have two contradictory beliefs at the same time, would harm the explanatory work of beliefs.

Conclusion 

Tamar Gendler has supplied a new concept of mental states which content is a package of representational, affective and motor signals. This concept is explanatorily valuable in cases of belief-behavior mismatch, in which a person consciously believes a proposition to be true, but in some way fails to act in accordance with it. I introduced two different situations in which a belief-behavior mismatch occurs. One in which appearances override evidence, meaning that even though the subject has an evidence-based belief, her behavior tells otherwise. The other concept is one in which a subject believes P to be true, but acting on habit she behaves as if not-P is true. I showed that our familiar cognitive attitudes belief and imagination cannot explain why subjects sometimes act opposed to their behavior. That is because it is contradictory to state the subject has a belief of P but simultaneously a belief of not-P. If a subject believes that P, she holds that P is true and it is impossible that she also holds that not-P is true. To say that the subject is merely imagining not-P to be true would be unnatural, because her behavior clearly shows that she really represents not-P to be true. Declaring the representation of not-P as an alief would explain this behavior. The subject associates an object or situation with an experience she has undergone. This leads to the occurrence of an alief, which contains a mixture of representation, affect and motor routine. When this association is made it directly activates a certain behavior and affect, because the content of an alief has both a behavioral and an affective component. Considering cases of belief-behavior mismatch we can say that a subject believes P and has an alief which represents not-P, even though I think the criticism of Eric Mandelbaum is legitimate. He is not showing successfully that ‘belief’ is not evidence-sensitive. Therefore the difference of alief and belief still lies in their nature of their sensitivity to evidence. Given that Tamar Gendler herself points out that the concept of aliefs is not fully worked out, it still has potential to be an extremely promising way of explaining our dissonant behavior. 

[1] Gendler states the following in her paper: “(a) the representation of some object or concept or situation or circumstance, perhaps propositionally, perhaps non-propositionally, perhaps conceptually, perhaps non-conceptually” (Gendler, 2007). 

[2] This is at least what Tamar Gendler holds. The discussion of the definition of belief is big, but in this paper I will take Gendler’s definition as sufficient. 

[3] Gendler emphasizes that not every alief necessarily has all of the three components. 

Literature

Gendler, T. 2007, "Alief and Belief" in Journal of philosophy ,105 (10): 634-663.

Fodor, J.A. 1983, The Modularity of Mind. Cambridge, Mass: MIT Press.

Mandelbaum, E. 2013, "Against Alief" in Philosophical Studies, 165 (1): 197-211.