Belief

• 1. Introduction

  ›

  Belief begins with snow. To believe that snow is white is to accept the truth of the proposition "snow is white." Yet few people ever pause to examine such a thought. Almost no one carefully considers whether the sun will rise tomorrow. We simply assume that it will. This is the strange territory that philosophers of belief have mapped for centuries. A belief is a subjective attitude that something is true, or that a state of affairs is the case. But what does it mean to hold an attitude you are not even thinking about? In epistemology, the term refers to attitudes about the world that can be either true or false. From here, harder questions follow. What is the rational way to revise a belief when new evidence arrives? Are beliefs fixed all-or-nothing things, or do they come in degrees? Must a belief be expressible in language, or can some beliefs have no words at all? Some thinkers go further still and argue that beliefs, as we imagine them, do not exist.

• 2. Rival Conceptions

  ›

  Representationalism is the traditionally dominant position, and it casts beliefs as attitudes toward representations, typically associated with propositions. On this view, those attitudes are part of the internal constitution of the mind that holds them. Beliefs form a special class of mental representation, because unlike perceptions or episodic memories they carry no sensory qualities. This makes it natural to treat them as propositional attitudes, distinguished from desires by their mode. A belief has what philosophers call a mind-to-world direction of fit. It tries to represent the world as it is, rather than trying to change it.

  Rahul believes it will be sunny today, and his attitude affirms that the proposition is true. Sofia desires that it will be sunny today, yet both are directed at the very same proposition. The difference shows when a forecast of bad weather arrives. Rahul is likely to revise his belief, while Sofia's desire stands unchanged. This tendency to revise upon new evidence is why beliefs are said to aim at truth.

  Functionalism breaks from this picture by defining beliefs not by the mind's internal makeup but by their causal role. A hard drive is the analogy: it is defined by the function of storing and retrieving digital data, whether built of plastic or steel, whether using magnetism or laser. In the same way, whatever is caused by perceptions in a certain way and causes behavior in a certain way counts as a belief. Seeing a traffic light turn red causes the belief that the light is red, which brings the driver's car to a halt. By this logic the belief might be ascribed to animals, hypothetical aliens, computers, even a self-driving car that behaves like a human driver.

  Dispositionalism is sometimes seen as a specific form of functionalism, tying a belief to a disposition to behave. A belief that there is a pie in the pantry is linked to affirming it when asked and going to the pantry when hungry. The trouble is that the mechanisms shaping behavior are too tangled to isolate any single belief. One might stay silent to keep the pie secret, or refuse to eat it while hungry because one also believes it is poisoned.

• 3. Belief As Interpretation

  ›

  Daniel Dennett argues that we ascribe beliefs to entities in order to predict how they will behave. He distinguishes the physical stance and the design stance, used for entities with simple behavioral patterns, from the intentional stance, applied to entities with more complex behavior. Take a chess player who moves her queen to f7. We predict the move by ascribing to her the desire to win and the belief that this move achieves it. The same procedure works on a chess computer. On this view, having a belief is relative to an interpretation, because there may be several equally good ways of ascribing beliefs to predict the same behavior.

  Donald Davidson offers another version through the thought experiment of radical interpretation. Here the goal is to make sense of another person's behavior and language from scratch, with no prior knowledge of their tongue. The process means ascribing beliefs and desires to the speaker, and the speaker really has them if the project can succeed in principle.

  Eliminativists, by contrast, hold that strictly speaking there are no beliefs at all. Instrumentalists agree there are none, yet add that belief-ascriptions remain useful, because they help us predict how entities will behave. Interpretationism can be read in this deflationary spirit, or in a more realistic one in which entities genuinely have the beliefs ascribed to them. For the realistic reading to hold, interpretationism may have to be defined as a methodology rather than as a claim about what truly exists.

• 4. Active And Graded

  ›

  An occurrent belief is one you are entertaining right now, such as actively thinking that the Grand Canyon is in Arizona. The great majority of beliefs are not active most of the time. They are merely dispositional, waking up when needed and sinking back afterward. The belief that 57 is greater than 14 was probably dormant in the reader's mind, became active on reading it, and will soon settle again. This split is sometimes equated with the divide between conscious and unconscious beliefs, but the two do not fully match. A belief can guide behavior and reasoning without the subject being conscious of it, which makes it an unconscious occurrent state.

  Formal epistemology hosts a deeper dispute about whether beliefs are full or partial. Full beliefs are all-or-nothing: either Pedro believes the Earth is bigger than the Moon or he does not. Yet some comparisons resist this frame, such as Pedro being more certain the Earth is bigger than the Moon than that it is bigger than Venus. These cases call for partial beliefs, or credences, formalized by numbers between 0 and 1. A degree of 1 is absolute certainty, a degree of 0 absolute disbelief, and the numbers between mark intermediate confidence. In the Bayesian approach these are subjective probabilities. A belief of degree 0.9 that it will rain tomorrow means the agent puts the chance of rain at 90 percent.

  The Lockean thesis says partial beliefs are basic, and a full belief is simply a partial belief above a chosen threshold, perhaps every belief above 0.9. Others reverse the order, treating a partial belief of degree 0.9 as a full belief that the probability is 90 percent. A third route drops probability entirely and measures degrees of disposition to revise a full belief, so a 0.6 belief and a 0.9 belief are both full, differing only in how easily new evidence can shift them.

• 5. What Beliefs Are About

  ›

  Lois Lane believes that Superman is strong, yet she does not believe that Clark Kent is strong, even though the two names refer to the same person. In ordinary contexts, swapping co-referring terms leaves a sentence's truth-value untouched, so "Superman is strong" and "Clark Kent is strong" stand or fall together. Belief ascriptions are more delicate. Where substitution fails, the belief is de dicto, set in a referentially opaque context. Where it succeeds, the belief is de re and the context is referentially transparent. In a de re sense Lois does believe Clark Kent is strong, while in a de dicto sense she does not.

  Mei and Benjamin both affirm that Jupiter is a planet, and the simplest reading, favored by atomists, is that they share one belief. Then Mei also believes Pluto is a planet, which Benjamin denies, suggesting their concepts of planet differ. Molecularists and holists conclude that the content of the Jupiter-belief depends on the Pluto-belief. Atomists deny such dependence, molecularists allow it among a few related beliefs, and holists let it hold between any two beliefs at all. W. V. Quine's confirmational holism feeds this view, arguing that confirmation happens at the level of a whole theory rather than for individual hypotheses. One cost is that genuine disagreement becomes rare, since disputants rarely share the exact web of beliefs needed to fix a shared point of dispute.

  Hilary Putnam presses the question of content outward with his twin Earth thought experiment. He imagines a twin Earth, identical to ours except that its water has a different chemical composition while behaving just like ours. The reader's thought that water is wet is about our water; the twin's identical thought is about theirs, despite the two readers sharing the same molecular makeup. Internalists hold that a person and a molecule-by-molecule copy share all the same beliefs. Externalists answer that the difference in content needs the environment to explain it, though critics note the two readers act in exactly the same way.

• 6. Knowledge And Its Cracks

  ›

  Plato's dialogue Theaetetus is where the trouble starts, the place where Socrates departs most clearly from the sophists who defined knowledge as justified true belief. Socrates dismisses the move from common opinion, doxa, to knowledge, episteme, refusing to grant knowledge even when an opinion is correct and backed by justification, by logos. Plato has been credited with the justified true belief theory, even though in the Theaetetus he elegantly dismisses it, and casts the argument as a cause of Socrates' death penalty.

  The definition itself gained approval during the Enlightenment, with "justified" set against "revealed." To know that a proposition is true, one must believe the true proposition and also have justification for the belief. That account suffered a setback with Gettier problems, situations that seem to meet every condition yet where many philosophers deny that anything is known. Robert Nozick proposed a fix, requiring that were the justification false, the knowledge would be false too. Bernecker and Dretske, writing in 2000, claimed no epistemologist since Gettier has seriously and successfully defended the traditional view. Paul Boghossian, in 2006, still saw the justified-true-belief account as the standard, widely accepted Platonic definition of knowledge.

  Three Ancient Greek concepts frame this older terrain. Pistis points to trust and confidence, doxa to opinion and acceptance, and dogma to the positions of a philosopher or a school such as Stoicism. The same vocabulary that built the theory of knowledge also marked the line between belief that something is so and belief in something. Belief-that is a propositional attitude to a claim that is true or false. Belief-in is closer to trust or faith and usually attaches to persons, central to religious traditions where belief in God is a chief virtue. The line blurs, since a belief in fairies can be read as a belief that fairies exist, and a belief in marriage as a belief that marriage is good.

• 7. Systems And Faith

  ›

  Jonathan Glover, following Meadows, insists that beliefs are always part of a belief system, and that no belief exists in isolation in the believer's mind. He pictures a patient told that the prescribed medicine is not working. At that moment the patient has wide latitude, free to believe the doctor is incompetent, that an assistant erred, that the patient's own body is unique, that Western medicine is ineffective, or even that Western science cannot discover truths about ailments. Glover holds that any belief can be kept if one really wants to, with help from ad hoc hypotheses, one belief held fixed while others shift around it.

  Rene Descartes offers Glover his sharpest example. "He starts off with the characteristic beliefs of a 17th-century Frenchman; he then junks the lot, he rebuilds the system, and somehow it looks a lot like the beliefs of a 17th-century Frenchman." To Glover, belief systems are not like houses but like boats: "Maybe the whole thing needs rebuilding, but inevitably at any point you have to keep enough of it intact to keep floating." He adds that people first realize beliefs can change, and may be contingent on upbringing, around age 12 or 15. Stephen Law goes blunter, calling belief in homeopathy, psychic powers, and alien abduction "claptrap" that can draw people in and hold them captive.

  Religious beliefs, unlike most other belief systems, are usually codified. Surveys often find that official doctrine and the privately held beliefs of members do not always agree. The term "fundamentalism" was first self-applied by anti-modernist Protestants in the United States, marking strict adherence to an interpretation of scripture. "Orthodoxy" first arose in Early Christianity, where the prevailing authority was the communion of bishops, often called the Magisterium, with "heterodox" as its antonym. Reform Judaism and Liberal Christianity show religious groups trying to fold Enlightenment ideals of rationality, equality, and individual liberty into their systems.

  Exclusivists treat other faiths as error, corruption, or counterfeit, a stance common among smaller new religious movements claiming a unique revelation. Some pair this with proselytization, strong in the Christian Great Commission and less emphasized in Islam, where the Quranic edict "There shall be no compulsion in religion" is often quoted. Inclusivists recognize some truth in all faith systems, an attitude the Baha'i Faith holds as doctrine. Pluralists make no distinction between faith systems, while syncretists blend traditions into a fusion, exemplified by Unitarian Universalism.

• 8. The Mind That Believes

  ›

  Abraham Maslow's research after World War II found that Holocaust survivors tended to be those who held strong religious beliefs, not necessarily through temple attendance, suggesting belief helped people cope in extreme circumstances. Viktor Frankl's Man's Search for Meaning details his own experience of religion's importance in surviving the Holocaust. The study suggested humans may need religious ideas to serve emotional needs: to feel loved, to belong to homogeneous groups, to have understandable explanations, and to be assured of ultimate justice. Critics counter that selecting subjects by religion may have introduced bias, and that all subjects being Holocaust survivors may have skewed results. Larson and colleagues, in 2000, called for more longitudinal research with better multidimensional measures.

  Lynne Rudder Baker, in her book Saving Belief, outlines four contemporary approaches to belief. The first holds our common-sense understanding is correct, a mental sentence theory defended by Jerry Fodor. The second, argued by Stephen Stich, holds it is not entirely correct but close enough for useful predictions. The third is eliminativism, proposed by Paul and Patricia Churchland, who compare the concept of belief to the four humours theory of medicine or the phlogiston theory of combustion, doomed to be discarded as neuroscience advances. The fourth, held by Dennett and Baker, calls belief scientifically invalid yet a successful predictive device, the intentional stance.

  Belief revision describes the modification of beliefs, often modeled by Bayesian updating, prized for its mathematical basis and conceptual simplicity. Yet such a process may not represent people whose beliefs are not easily cast as probabilistic. Persuasion is the broad umbrella for changing others' beliefs, taking forms such as consciousness raising in activist or political contexts. Resistance depends on more than evidence, including a message's credibility, social pressures, and anticipated consequences. Albert Einstein is often quoted as saying "Common sense is the collection of prejudices acquired by age eighteen," and indeed political beliefs depend most strongly on those common where one lives, while most people keep the religion taught in childhood. Physical trauma, especially to the head, can radically alter what a person believes. In the DSM-5, delusions are defined as fixed false beliefs that do not change even when confronted with conflicting evidence, the point at which belief, the most ordinary thing in the mind, becomes a clinical concern.