Hello! Welcome back to our study of epistemology.
In our last two lessons, we explored foundationalism, the theory that our beliefs are justified like a pyramid, with a strong base of "properly basic beliefs" supporting everything else. We saw how classical foundationalists demanded an infallible foundation (self-evident truths, incorrigible beliefs), while modern foundationalists like Alvin Plantinga argued for a broader, more fallible foundation that could include perception, memory, and even belief in God.
Today, we investigate a radically different answer to the epistemic regress problem. What if our belief system isn't a pyramid at all? What if it's more like a raft, where all the pieces hold each other afloat with no single piece serving as the ultimate foundation?
This brings us to our learning outcome: to describe coherentism as an alternative response, explaining how a web of beliefs can be mutually supporting. We will explore this "web of belief" model, see how justification is thought to emerge from it, and consider some of its most significant challenges.
From Pyramid to Web: The Coherentist Picture
Foundationalism's response to the regress problem is to stop the chain of "why?" questions with a foundational belief that needs no further justification. Coherentism's response is to argue that the "chain" metaphor is wrong from the start. Justification isn't a linear process; it's holistic. A belief is justified not because it rests on an unshakable base, but because it fits into and is supported by a whole network of other beliefs.
The best way to get a feel for this is through the classic analogy of a spider's web.
Epistemology: Foundationalism & Coherentism
This clip from the channel 'Groundwork For The Metaphysics of Monkeys' provides an excellent introduction to coherentism using the spider web analogy.
Please watch from 15:31 to 16:56. Notice how justification isn't about a single thread, but about the strength of the connections within the entire system.
As the video explains, a single strand in a web is weak, but when woven together, they create a strong, durable network. For a coherentist, the justification for any single belief works the same way. Its strength comes from its connections to all the other beliefs in the system.
The Mechanics of Mutual Support
Saying beliefs "cohere" or are "mutually supporting" sounds good, but what does it actually mean? Justification can't be a mysterious property that just appears. Coherentists point to several concrete relationships between beliefs that create a strong, coherent web.
To understand these, let's turn to a more detailed textual explanation.
The OpenStax textbook 'Introduction to Philosophy' offers a fantastic chapter on justification. The section on Coherentism clearly explains the theory and how it contrasts with foundationalism.
Please read the section titled 'Coherence.' It starts right after the section on Foundationalism. Focus on two things: The difference between mere 'consistency' and active 'support.' The example of trying to remember where your friend Faruq is from. See how multiple, separate beliefs reinforce each other to strengthen a conclusion.
As this reading makes clear, coherence is more than just a lack of contradiction. My beliefs that "it is Tuesday" and "grass is green" are consistent, but they don't support each other in any meaningful way. For a system to be coherent, the beliefs must actively work together. The most important kinds of relationships that build coherence are:
-
Logical Consistency: This is the absolute minimum. A system of beliefs that contains contradictions (e.g., believing "All birds can fly" and also "Penguins are birds that cannot fly") is incoherent.
-
Explanatory Relations: This is the heart of coherence. Beliefs in a coherent system explain one another. In the "Faruq from Tennessee" example, the belief that he is from Tennessee provides the best explanation for a whole cluster of other beliefs: his hat, his car sticker, his accent, and his stories about the Smoky Mountains. In turn, those beliefs make the belief "he is from Tennessee" much more likely. They create a tight, mutually reinforcing explanatory loop.
-
Inductive and Deductive Relations: Beliefs can also support each other through logical and probabilistic inference. If I believe "All men are mortal" and "Socrates is a man," these two beliefs work together to support the belief "Socrates is mortal."
For a deeper dive into these components, the Internet Encyclopedia of Philosophy offers a more technical breakdown.
For those interested in the finer details, this article on 'Coherentism in Epistemology' unpacks the components of the coherence relation itself.
Please read the section 'b. What is the Makeup of the Coherence Relation?'. You don't need to master every detail, but notice how philosophers break down coherence into specific propositional relations (deductive, inductive, explanatory) and a psychological condition (the beliefs must actually be held and related in the believer's mind).
Test your understanding!
Imagine a detective investigating a murder. She has the following beliefs:
A. The suspect's fingerprints are on the murder weapon.
B. A witness saw the suspect arguing with the victim an hour before the murder.
C. The suspect had a strong financial motive.
D. The suspect claims he was at the cinema across town at the time of the murder.
From a coherentist perspective, which belief weakens the justification of the belief system? Why?
Show answer
Belief D, the suspect's alibi, weakens the system. Beliefs A, B, and C are highly coherent: they are consistent and provide a strong explanatory web supporting the conclusion that the suspect is guilty. Belief D introduces a contradiction. It does not cohere with the other beliefs. The detective's job, from a coherentist standpoint, is to determine whether there's a way to resolve this incoherence—perhaps the alibi is false, or perhaps her other beliefs are misleading. Until this is resolved, her overall justification is weakened.
Two Major Obstacles for Coherentism
While the web metaphor is elegant, coherentism faces some powerful objections. Any description of the theory is incomplete without them.
1. The Isolation Objection
A set of beliefs can be perfectly coherent—internally consistent, with rich explanatory connections—and yet be completely disconnected from reality. A well-written fantasy series like The Lord of the Rings has a highly coherent set of "beliefs" within its narrative, but that doesn't make it true. How can coherence alone justify a belief if the entire system might be a fiction?
The OpenStax article you read earlier presents this objection with the excellent thought experiment of "Dinah," a person trapped in a detailed virtual reality. Her beliefs are all coherent, but they are all false. This suggests that for a belief system to constitute knowledge, it needs to be "anchored" to the external world in some way—a point that foundationalists are quick to emphasize.
2. The Alternative Systems Objection
This is a variation of the isolation objection. If coherence is the only source of justification, what do we do when we have two or more competing systems of belief that are both internally coherent, but are mutually exclusive?
Epistemology: Foundationalism & Coherentism
Let's return to the 'Groundwork' video, which raises this exact point using an example that is highly relevant to your interests.
Please watch from 16:56 to 18:48. The speaker brings up the idea of Christianity and science as two different, internally coherent systems. This highlights the problem: if both are coherent, but they conflict, how do we choose between them? Coherence alone doesn't seem to give us the tools to do so.
This is a profound challenge. As the speaker notes, if you find yourself automatically dismissing one coherent system in favor of another, you must ask why. Your reason for choosing probably appeals to some standard outside of mere coherence (e.g., "science connects with empirical evidence," which is a foundationalist-style appeal). This "scream in the mind" suggests that coherence, by itself, might not be enough for justification.
Conclusion
Today we have sketched out coherentism, a major internalist theory of justification and a direct rival to foundationalism. Instead of a pyramid built on a foundation, it presents our knowledge as a self-supporting web.
Key Takeaways:
- Coherentism responds to the regress problem by adopting a holistic view of justification, where individual beliefs are justified by their membership in a coherent system.
- The metaphor for this structure is a web or raft, where beliefs are mutually supporting, not linearly supported by a foundation.
- Coherence is built from relationships like logical consistency, explanatory power, and inferential connections.
- The theory's primary weaknesses are the Isolation Objection (a coherent system could be pure fiction) and the Alternative Systems Objection (coherence alone doesn't help us choose between two conflicting but coherent systems).
Preview of the Next Lesson:
Both foundationalism and coherentism are internalist theories. They locate justification in factors that are "internal" to you as a believer: your access to your own beliefs, your sensory experiences, and the logical relations between your ideas.
In our next lesson, we will make a major shift and explore externalism. We'll start with the most prominent externalist theory, process reliabilism, which argues that a belief is justified if it's produced by a reliable cognitive process, even if you have no internal access to or understanding of why that process is reliable. This will challenge many of our intuitions about what it means for a belief to be justified.