Belief and Credence by Elizabeth Grace Jackson

Paper and the Author:

The Aim of the Paper:
In “Belief and credence: why the attitude-type matters,” Jackson focuses on five epistemological debates and examines how each position (belief-first, credence-first, or dualism) influences one’s views on those debates. She does not argue for one or another position but sheds light on the significance of the relationship between belief and credence in shaping these debates. Jackson suggests that understanding whether one holds a belief-first, credence-first, or dualist perspective is crucial because it constrains and influences how it relates to issues like permissivism, disagreement, pragmatic encroachment, doxastic voluntarism, and the relationship between doxastic attitudes and rational action.

Why does it matter?
The relationship between belief and credence is a central issue for epistemology because it changes our understanding of epistemology depending on how we understand the relationship between belief and credence.

Elizabeth Grace Jackson is a Philosophy Professor at Saint Louis University (as of June 2024)
Here is the full article: Belief and credence: why the attitude-type matters

Background and Key Concepts:

Note that some authors may have different conceptions or definitions for each concept. The conception of the following concepts pertains to the paper.

Belief is an attitude where a person takes some proposition to be the case or represents it as true. Beliefs are categorical; you either believe a proposition or you do not​​.

Credence refers to a graded or probabilistic belief. It represents the degree of confidence in a proposition (sentences with truth values). Unlike categorical belief, credences can vary in intensity and represent the subjective probability that a proposition is true​​.

Permissivism is the view that a body of evidence can rationally permit more than one doxastic response (i.e., more than one rational belief or credence). For instance, two people with the same evidence might rationally conclude differently​​. This can be broken down into credence permissivism and belief permissivism.

Uniqueness Thesis: The idea that a particular body of evidence justifies at most one proposition. This stands in contrast to permissivism. This can be broken down into credence uniqueness and belief uniqueness.

Disagreement: This concept involves how individuals should respond when they encounter epistemic peers (people with equal evidence and cognitive abilities) who disagree with them.

The opposing positions are the following:

• Steadfastness: This position argues that peer disagreement does not necessarily warrant changing one’s beliefs. It claims it is often rational to remain steadfast in one’s beliefs even when faced with peer disagreement. This position recognizes that we frequently encounter disagreements, like in political or religious matters, where it seems rational to maintain our convictions despite opposing views.
• Conciliationism: This position maintains that disagreement with an epistemic peer is a reason to alter one’s doxastic attitudes. For example, suppose two equally competent mathematicians reach different conclusions about the same problem. In that case, each should reconsider their stance and possibly withhold belief or lower their confidence in their initial decision.​

Pragmatic Encroachment: This debate centers on whether practical interests can influence the epistemic justification of a belief. Pragmatists argue that practical stakes can affect what one is justified in believing, whereas non-pragmatists (or purists) deny this and maintain that epistemic justification is independent of practical interests​​.

Then we have the following positions:

• Credal Pragmatism
• Belief Pragmatism
• Credal Purism
• Belief Purism

Doxastic Voluntarism: The view regarding whether individuals have voluntary control over their beliefs. Some argue that we can choose our beliefs, while others contend that belief formation is involuntary​​. Voluntarists argue that individuals can directly control their beliefs and credences, similar to how one might control physical actions like raising a hand​. Involuntarists deny this, asserting that doxastic attitudes cannot be controlled directly. They argue that beliefs and credences are involuntary responses to evidence​

Then we have the following positions:

• Belief voluntarism
• Credence voluntarism
• Belief involuntarism
• Credence involuntarism

Belief-First vs. Credence-First vs. Dualism: These are three views on the relationship between belief and credence:

• Belief-First holds that belief is the fundamental attitude, and credence is only a belief with probabilistic content (e.g., “the probability that it will rain tomorrow is 0.7”). Therefore, credence and non-credence are beliefs that differ in content but must be true or false. For credence (beliefs with probabilistic content), a proposition “the probability that it will rain tomorrow is 0.7” is true if, in fact, the “probability that it will rain tomorrow is 0.7.” Similarly, our belief is true if we believe p with probabilistic content and p accurately reflects the probabilistic fact.
• Credence-First holds that credences are fundamental attitudes, and beliefs are a form of high credence above some threshold.

So what is going on above is a reduction. Credence-first reduces beliefs to credence, while Belief-First reduces credence to beliefs. Each has a single distinct attitude.

• Dualism: Both belief and credence are fundamental and irreducible to one another​. This a popular view today.

Models of Rational Action:

• Decision Theory: This model asserts that an action is rational if it maximizes expected value. It involves calculating probabilities (credences) and utilities​.
• Belief-Desire Model: This model posits that an action is rational if it aligns with one’s beliefs and desires. It involves taking beliefs and desires as inputs​.

Paper Reconstruction:

Permissivism vs. Uniqueness

[Read Jackson’s examples]

• Example of permissivism where two people can rationally have different doxastic attitudes given the same evidence (belief in God example): Page 2480.
• Example explaining how credal permissivism and belief permissivism come apart: Page 2481.

Jackson does not explicitly do this, but based on her paper, I will track the number of positions about belief-first, credence-first, and dualism.

The Permissivism(see above for explanation) debate diverges into the following positions: Given a set of evidence and propositions, we can have multiple rational doxastic attitudes or only one. Here, we call the former Permissivism and the latter Uniqueness. For the latter, if two people believe something different, their evidence must differ. When evidence is identical, but beliefs diverge, Uniqueness concludes that at least one of the two is irrational.

Regarding the above debate, Credal Permissivism and Belief Permissivism could be possible. If each is false, Uniqueness will be true about credence and beliefs, respectively. If permissive is true respectively, two or more people can have different credences or beliefs given identical evidence.

As Jackson states, we can have Belief Permissivism false while holding that Credal Permissivism is true. We can hold p as true with identical evidence but with different confidence levels. Still, she argues that Credal Uniqueness is implausible because, given a set of evidence, one must arrive at a single rational credence. This is overdemanding since we often find different credences for a given body of evidence. If she is correct, we cannot hold Belief Permissivism as true while Credal Permissivism is false.

Supposing that Credal Permissisim is true, Credal Uniqueness is false. We can have Belief Permissivism while Credal Permissisism OR Belief Uniqueness while Credal Permissisism is True.

What are the implications if Belief-First, Credence-First, or Dualism is the correct theory?

We need to consider each relationship between belief and credence in combination with each possible stance on belief and credence, including permissivism/uniqueness.

This leads to the following combinations:

1. Belief-First and Belief Permissivism while Credal Permissivism
2. Belief-First and Belief Uniqueness while Credal Permissivism
3. Credence-First and Belief Permissivism while Credal Permissivism
4. Credence-First and Belief Uniqueness while Credal Permissivism
5. Dualism and Belief Permissivism while Credal Permissivism
6. Dualism and Belief Uniqueness while Credal Permissivism

Credal-Frist with Credal Permissivism reduces to Belief Permissivism

Thus, we can eliminate the fourth option above. Now, we have 5 options that we cannot eliminate so far.

Steadfast vs. Conciliationism

[Read Jackson’s examples]

• Example of disagreement where one should alter their doxastic attitudes (splitting the bill in a restaurant): Page 2482.
• Example where “splitting the difference” in credence may not be appropriate (doctor determining drug dosage): Page 2483.

One disagreement in epistemology concerns how individuals should respond to epistemic peer disagreement, whether they should change their beliefs or credence when faced with peer disagreement. There are opposing positions: steady vs. conciliationism (see background concepts above for detail).

Jackson argues that combining belief-first with conciliationism is challenging because beliefs are coarse-grained. Altering beliefs in response to disagreement is not straightforward because there are limited intermediate attitudes between belief and disbelief

Moreover, she argues that a credence-first view initially aligns well with conciliationism, as one can “split the difference” between differing credences. However, practical cases like determining the confidence level in prescribing drug dosages show that mechanical averaging might not always be appropriate, and sometimes increasing confidence is more rational despite disagreement.​

Dualism could hold steadfast about beliefs but conciliatory about credences and vice versa. It can also have only steadfast or conciliationism.

Therefore, conciliationism is prima facie incompatible with belief-first or credence-first view.

1. Belief-First and Belief Permissivism while Credal Permissivism, Steadfast
2. Belief-First and Belief Uniqueness while Credal Permissivism, Steadfast
3. Credence-First and Belief Permissivism while Credal Permissivism, Steadfast
4. Dualism and Belief Permissivism while Credal Permissivism, Steadfast
5. Dualism and Belief Permissivism while Credal Permissivism, conciliationism
6. Dualism and Belief Permissivism while Credal Permissivism, Steadfast & conciliationism
7. Dualism and Belief Uniqueness while Credal Permissivism, Steadfast
8. Dualism and Belief Uniqueness while Credal Permissivism, conciliationism
9. Dualism and Belief Uniqueness while Credal Permissivism, Steadfast & conciliationism

Pragmatism vs. Purism

[Read Jackson’s examples]

• Example illustrating how practical stakes affect epistemic justification (Hannah’s bank example): Page 2485.

The debate over pragmatic encroachment revolves around whether practical interests can affect epistemic rationality or justification of beliefs and credences.

Belief-first tends to align pragmatism or purism equally across both beliefs and credences. If one holds that beliefs are influenced by practical interests, then credences should be as well, and vice versa​.

Credence-first allows for credal purism and belief pragmatism, depending on whether practical factors can affect the threshold for belief. This view can accommodate varying thresholds for rational belief based on the stakes involved. However, whether we control changing the threshold is a different matter.

Dualism allows for independent combinations of pragmatism and purism for beliefs and credences. It offers an approach in which beliefs might be pragmatic in low-stakes scenarios and credences in high-stakes situations, each serving different epistemic purposes.​

If we take into account pragmatism vs. purism, we have 36 positions. We can drop 4 Belief-first views since pragmatism about belief implies Pragmatism about credence (and vice versa) and purism about belief implies Purism about credence (and vice versa).

Therefore, we now have 32 positions.

Voluntarism vs. Involuntarism

[Read Jackson’s examples]

• Example discussing direct control over doxastic attitudes (raising one’s hand analogy): Page 2487.
• Example where voluntarism about belief suggests voluntarism about credence: Page 2488.

The doxastic voluntarism debate focuses on whether we can have direct control over our doxastic attitudes (beliefs and credences).

The truth of credence-first depends on whether credal voluntarism is true. While the literature has primarily focused on beliefs, it remains an open question whether credences can be directly controlled.​

Since it is controversial credal voluntarism, Jackson supposes that it is false.

For belief-first, voluntarism about belief suggests voluntarism about credence, and vice versa, since beliefs with probabilistic content should be similarly controllable​ if beliefs are controllable.

Dualism allows for a separation where one might maintain voluntarism for beliefs but deny it for credence or vice versa. Dualism does not require the ability to directly control the threshold for belief, thus providing a more flexible approach​(Belief and credence_why…)​.

Before eliminating anything concerning voluntarism, we have 128 views. If we suppose credal voluntarism is implausible like Jackson (which I agree), we have 64 positions. When we also account for the restriction of belief-first, we have 60 positions.

Doxastic Attitudes and Rational Action

[Read Jackson’s examples]

• Example of rational action aligning with credences (decision theory vs. belief-desire model): Page 2489.
• Example illustrating rational action despite belief (skating on a frozen lake, brother missing): Page 2490.

Finally, Jackson discusses how belief and credence relate to models of rational action.

Credence-First aligns with decision theory, as credences are taken as inputs for maximizing expected value. However, psychological evidence shows that people often fail to maximize expected value due to cognitive limitations, questioning the practicality of this model as a guiding norm for all decision-making​.

Belief-First aligns with the belief-desire model, which might be better suited for guiding action given cognitive limitations. However, it faces challenges when rational action requires acting against one’s belief due to high stakes or probabilistic considerations​.

Dualism allows using beliefs in low-stakes situations and credences in high-stakes situations. By justifying beliefs and credences in different contexts​​, dualism attempts to meet the Bayesian Challenge. So, dualism offers a flexible approach to justifying different uses of belief and credence in varying contexts.

Critical Note:

A Focus on Dualism:

Given these steps, it is clear that the majority of the positions are focused on dualism due to the following reasons:

• Flexibility of Dualism: Dualism allows for various combinations of belief and credence stances. Since dualism posits that belief and credence are fundamental and irreducible, it naturally accommodates multiple combinations leading to more positions.
• Reduction in Belief-First and Credence-First Positions: Belief-first and credence-first positions are more restrictive. For example, belief-first views are reduced when considering pragmatism/purism and voluntarism, as these views imply straightforward connections between belief and credence, leading to fewer viable combinations.
• Robust Combinations in Dualism: Dualism allows for independent consideration of belief and credence in relation to permissivism, steadfastness, conciliationism, pragmatism, purism, and voluntarism. This independence leads to more possible combinations compared to belief-first or credence-first views.

Footnotes:

Reference:

• Jackson, E. G. (2019). Belief and credence: why the attitude-type matters. Philosophical Studies, 176(9), 2477–2496. https://doi.org/10.1007/s11098-018-1136-1

*Disclaimer: This article is a creative interpretation and synthesis work, drawing inspiration from multiple sources. While efforts have been made to ensure the accuracy of the information presented, readers are encouraged to consult the source material for a comprehensive understanding.

**Please note that I often write the Critical Note (constructive criticism) section after posting the article. If the section is left blank, it is intentional.

***I reserve the right to edit this page at my convenience.