See full article here.

Summary notes for The tripartite view

The distinction between acquaintance knowledge, ability knowledge and propositional knowledge.

Ability: knowledge ‘how’ – e.g. “I know how to swim”
Acquaintance: knowledge ‘of’ some person, place or thing through direct awareness – e.g. “I know London”. Knowing what it is like to have this direct awareness.
Propositional: knowledge ‘that’ – e.g. “I know that London is the capital of England”
Epistemologists are mainly concerned with propositional knowledge because it is the type which can be represented in language and communicated.
The reason for the distinction is that there seem to be types of knowledge other than propositional. E.g. it’s not possible to reduce knowledge of how to swim to propositions. A person can’t be taught how to swim through learning facts. Nonetheless, it seems to make sense to say that some people know how to swim. So, there must be more types of knowledge than merely propositional.

The nature of definition (including Linda Zagzebski)

Epistemology has to start with a definition of knowledge, which raises the question of what a definition is and how it functions.
Zagzebski explains that epistemologists seek what is called a ‘theoretical definition’. This is is opposed to a practical definition, meaning a common-sense ordinary way that a word might be used by the average person. Not attempting to have precise theoretical clarity.
A theoretical definition attempts clarity through providing a necessary truth for what a thing must be.
A necessary truth is one that must be true. A theoretical definition will thus provide a necessary truth for what knowledge must be. It will explain what must be the case in order for knowledge to be held. This will remove any ambiguity, since either the conditions are present and so knowledge is held, or they aren’t and knowledge isn’t held.
This is achieved through necessary and sufficient conditions. E.g. ‘a batchelor is an unmarried man’ is a theoretical definition because it has precise necessary and sufficient conditions. Being unmarried is a necessary condition but not sufficient. Being unmarried and being a man are the necessary and sufficient conditions for being a batchelor.
Epistemologists want to find the conditions which are necessary and sufficient for knowledge since they will have then found the necessary truth for what must obtain in order to have knowledge, and thus a theoretical definition of knowledge.
Zagzebski also claims that theoretical definitions can be either real or conceptual. A real definition defines something with reference to an essential property in the world. E.g. ‘water is H2O’. An conceptual definition is when something socially constructed is defined, e.g. ‘A bachelor is an unmarried man’. Zagzebski says that we don’t know whether knowledge will have a real or conceptual definition until we successfully define it.

How propositional knowledge may be analysed/defined

Epistemologists argue that we can analyse or test the success of theoretical definitions of knowledge through truth condition analysis. This involves proposing necessary and sufficient conditions for knowledge which are then tested through trying to think of counter-examples.
E.g. JTB is tested through trying to think of examples of knowledge without one of the conditions (testing the individual necessity of JTB) or cases where the conditions are present and yet knowledge is not (testing the joint sufficiency of JTB).
Zagzebski argues that even if we could think of a definition which no one can think of counterexamples to, that could just be because we lack imagination. After all, Plato’s definition of knowledge was accepted for 2500 years until Gettier wrote his paper. So, Zagzebski says we can’t ever really be absolutely certain of a definition of knowledge since our only method for analysing or testing it depends on our ability to think of counterexamples.

The tripartite view

Epistemologists hope to find a theoretical definition of propositional knowledge, which would provide a necessary truth for what knowledge must be.
A necessary truth can be defined through finding its necessary and sufficient conditions.
Individually necessary means that all of them are required for knowledge.
Jointly sufficient meaning that if you have all three you must have knowledge, nothing more is needed.
The tripartite theory of knowledge claims that: S knows p if and only if
S believes p
S has justification for p
p is true

The individual necessity of the conditions

Zagzebski says that knowledge involves ‘cognitive contact with reality’. So knowledge is a relation between a thinking mind and reality.
The tripartite view aims to satisfy this description with its three criteria.
Truth
Truth seems essential as knowledge that is false lacks cognitive contact. Because, a false proposition doesn’t actually represent reality, so they lack the right kind of contact with reality.
Belief
Belief is a mental representation of reality which could be either true or false. This is necessary for the ‘cognitive’ element of knowledge.
If a person’s mind doesn’t contain a representation or understanding of the knowledge claimed about reality, then they have not cognitive contact with reality.
Justification
Then a condition is required which explains how the belief relates to its truth in a non-accidental way.
This is the most controversial part of defining knowledge.
Justification seems to be a good candidate because it provides a reason for the truth of the belief and seems to avoid lucky true beliefs which would lack cognitive contact.
E.g., lucky lottery winner who claimed to know they would win.

The issue that the conditions are not individually necessary: knowledge without justification

JTB claims that its conditions are individually necessary.
This means that if one of the conditions is not present, then knowledge is not had. So, if there are examples of knowledge without truth, then truth is not a necessary condition for knowledge, so JTB would fail.
There seem to be cases of knowledge which are known without justification.
This includes the immediate objects of perception.
E.g., If I have an experience of blue
If I wanted to claim that there really was a blue object in an external world, that might require justification.
However it seems I can know that I am experiencing blue without needing justification.
Other more contentious examples would be mathematical knowledge, such as that 1+1=2, or knowing that I exist or that God exists.

The issue that the conditions are not individually necessary: knowledge without truth

JTB claims that its conditions are individually necessary.
This means that if one of the conditions is not present, then knowledge is not had. So, if there are examples of knowledge without truth, then JTB fails.
Scientific knowledge seems to be a case of having knowledge without the requirement of truth as a condition for having it. E.g. scientists used to believe in all sorts of things that they now consider false. E.g. scientists used to think eating eggs every day was unhealthy, but now think it is fine.
This is because science only ever provides us with what we currently have most reason to believe based on the available evidence. As scientists get more evidence, their views change. So, some of our current scientific knowledge will one day be shown to be false by new better evidence. However, the point is we still call it knowledge. Even if though it could be false, it is still called ‘scientific knowledge’.
So, truth is not a necessary condition for being knowledge. The scientific definition of knowledge would be justified belief, where justification is the use of the scientific method.

The issue that the conditions are not individually necessary: knowledge without belief

JTB claims that its conditions are individually necessary.
This means that if one of the conditions is not present, then knowledge is not had. So, if there are examples of knowledge without belief, then JTB fails.
Imagine a student taking their exam. They revised something but then forgot it. Yet in the exam, when they read the question, the answer they had revised pops into their mind because they had revised it. They don’t remember revising it, they don’t know why it popped into their mind, but they decide to just go with it and write it down. It looks like they know the answer, it wasn’t just a lucky random guess. However, they forgot that they knew it, and this meant they didn’t really believe it. If you were to ask them whether they believed it was the right answer, they would say no, they just wrote it because it popped into their mind. However, they had revised it, they produced the answer, so it seems to make sense to say that they knew it.
Therefore: Theoretical definitions of knowledge attempt to provide a necessary truth about what knowledge must be in all cases. A single counter-example is sufficient to show that this fails. Cases of knowledge without belief therefore aim to show that any theoretical definition of knowledge with belief as a criteria fails.

The issue that the conditions are not sufficient – Gettier’s original two counter examples

Gettier claims to have found counter-examples to the joint sufficiency of the tripartite definition.
Case 1:
Smith is at a job interview and has justification for believing (due to being told by the manager) that Jones will get the job. He also has justification for thinking Jones has 10 coins in his pocket because he saw Jones counting them. So, Smith forms the belief “the person who will get the job has 10 coins in their pocket”.
It turns out Smith does get the job and by luck happens to have 10 coins in his pocket.
So, Smith had a justified true belief.
Case 2:
A disjunctive proposition involves the logical operator ‘or’. It is true if either or both of the disjuncts are true.
Smith has justification for thinking Jones owns a Ford car. Smith then thinks of the disjunctive proposition “Jones owns a Ford or Brown is in Barcelona”.
By luck, Jones actually does not own a Ford.
Also by luck, Brown actually is in barcelona.
So, the proposition is true, because it’s true that Brown is in Barcelona. Smith has justification, for believing the proposition is true, but it turned out to be true by luck for different reasons (because of the other disjunct being true).
In both these Gettier cases, we would not want to say Smith had knowledge because it was only by luck that Smith had a justified true belief. So, it lacked cognitive contact with reality.
Gettier has found cases where the tripartite criteria are present, yet knowledge is not.
So, Gettier problems are counterexamples to the joint sufficiency of the tripartite criteria.
It cannot be a correct theoretical definition of knowledge about what knowledge necessarily is if there are cases where the criteria obtain but knowledge does not.

Infallibilism

Epistemologists hope to find a theoretical definition of knowledge, which would provide a necessary truth for what knowledge must be.
This is achieved by finding criteria that are individually necessary and jointly sufficient.
Individually necessary means that all of them are required for knowledge.
Jointly sufficient meaning that if you have all three you must have knowledge, nothing more is needed.
Infallibilism is a variant of the tripartite view.
It claims that S knows p if and only if:
S believes p
p is true
S has infallible justification for believing p

How Infallibilism solves Gettier cases

This definition is meant to avoid the problem of epistemic luck by requiring that justification provide indubitable support for the truth of the belief.
In Gettier’s original counter-examples, Smith has justification for his true belief, but not infallible justification.
The manager could have been lying or deluded when telling Smith that Jones would get the job. Smith could have been hallucinating when seeing Jones driving a Ford.
So the infallibilist definition is meant to avoid Gettier problems by explaining how Smith does not have knowledge.

No false lemmas

Epistemologists hope to find a theoretical definition of knowledge, which would provide a necessary truth for what knowledge must be.
This is achieved by finding criteria that are individually necessary and jointly sufficient.
Individually necessary means that all of them are required for knowledge.
Jointly sufficient meaning that if you have all three you must have knowledge, nothing more is needed.
A false lemma is a false belief from which a conclusion is inferred.
No false lemmas theory claims that knowledge S knows p if and only if:
S believes p
p is true
S has justification for believing p
S’s belief that P is not inferred from any false beliefs

How NFL solves gettier cases

Sometimes false lemmas can be used to infer true conclusions due to luck. This happened in Gettier’s original counter examples. Making no false lemmas a condition of knowledge is meant to avoid lucky true beliefs counting as knowledge.
E.g. Smith used the false belief that Jones would get the job to come to his conclusion that the person who will get the job has 10 coins in their pocket. The belief that jones owns a ford was also a false belief used to infer the conclusion that ‘either jones owns a ford or brown is in barcelona’.

Reliabilism

Epistemologists hope to find a theoretical definition of knowledge, which would provide a necessary truth for what knowledge must be.
This is achieved by finding criteria that are individually necessary and jointly sufficient.
Individually necessary means that all of them are required for knowledge.
Jointly sufficient meaning that if you have all three you must have knowledge, nothing more is needed.
Reliabilism claims that knowledge S knows p if and only if:
S believes p
p is true
S’s belief p was caused by a reliable cognitive process
A reliable cognitive process is a mental faculty which causes a high percentage of true beliefs, such as memory or perception.

How reliabilism solves Gettier cases.

Reliabilism is meant to solve Gettier cases and explain why lucky true beliefs are not knowledge, by removing the justification criterion.
The problem with justification is that the reason for the truth of a belief and the actual justification a person has can come apart. This opens a space for epistemic luck.
Smith relied on justification to reach his conclusion, either that the manager told him Jones would get the job, or seeing Jones driving a ford.
However the actual reason for the truth of the belief was that he got the job, or that Brown was in Barcelona
Reliabilism solves this by requiring that the truth of the belief be caused by the third criteria.
If Smith’s true belief were instead caused by a reliable cognitive process, then it couldn’t have been caused by luck. So the reliabilist definition is meant to avoid epistemic luck by causally relating the conditions so as to exclude luck as a cause of the true belief.

Virtue epistemology

Epistemologists hope to find a theoretical definition of knowledge, which would provide a necessary truth for what knowledge must be.
This is achieved by finding criteria that are individually necessary and jointly sufficient.
Individually necessary means that all of them are required for knowledge.
Jointly sufficient meaning that if you have all three you must have knowledge, nothing more is needed.
Virtue epistemology claims that s knows p if and only if:
S believes P
P is true
S’ belief that P is true was caused by their exercising of intellectual virtue.
Intellectual virtue is any character trait that would incline someone to gaining true beliefs- e.g. attention to detail, desire for truth, self-awareness of bias.
The virtue epistemologist claims that knowledge occurs when a person’s intellectual virtue causes a true belief.
Knowledge is only had when the (e.g. attention to detail & desire for truth) causes a true belief.
E.g. a judge – convicts someone as guilty because they don’t like the way they look – imagine the person actually was guilty – their guilt would be a true belief – but not caused by intellectual virtue, so not knowledge.
If a judge considered the evidence with attention and lack of bias however – and that caused their judgement of not guilty – their true belief would count as knowledge.

How virtue epistemology solves Gettier cases.

Virtue epistemology is meant to solve Gettier cases and explain why lucky true beliefs are not knowledge, by removing the justification criterion.
The problem with justification is that the reason for the truth of a belief and the actual justification a person has can come apart. This opens a space for epistemic luck.
Smith relied on justification to reach his conclusion, either that the manager told him Jones would get the job, or seeing Jones driving a ford.
However the actual reason for the truth of the belief was that he got the job, or that Brown was in Barcelona
Virtue epistemology solves this by requiring that the truth of the belief be caused by the third criteria.
If Smith’s true belief were instead caused by the exercise of intellectual virtue then it couldn’t have been caused by luck. So the Virtue epistemology definition is meant to avoid epistemic luck by causally relating the conditions so as to exclude luck as a cause of the true belief.

The tripartite view model essay plan

Note that this model essay plan is merely one possible way to write an essay on this topic.

Points highlighted in light blue are integration points
Points highlighted in green are weighting points

Intro:

This essay will argue that…
Epistemologists hope to find a theoretical definition of knowledge, which would provide a necessary truth for what knowledge must be.
A necessary truth can be defined through finding its individually necessary and jointly sufficient conditions.
Individually necessary conditions are those which are each required for knowledge.
Jointly sufficient conditions are those for which if you have them then you must have knowledge, nothing more could be is needed.
The tripartite theory of knowledge claims that: S knows P if and only if
S believes P
S has justification for P
P is true
The strength of this theory is how intuitive its criteria account for what Zagzebski called knowledge’s ‘cognitive contact with reality’.
Truth seems essential as knowledge that is false lacks contact with reality.
Belief accounts for the ‘cognitive element, as beliefs are a mental representation of reality which could be either true or false.
A condition is then required which explains how the belief relates to its truth in a non-accidental way.
We wouldn’t intuitively think a true belief were knowledge if it were a lucky guess, for example.
Which criteria to use here is one of the most controversial issues in defining knowledge.
Justification seems to be a good candidate because it provides a reason for the truth of the belief and seems to avoid lucky true beliefs which would lack cognitive contact.

JTB, Gettier & Infallibilism

Gettier claims to have found counter-examples to the joint sufficiency of the tripartite definition.
Case 1:
Smith is at a job interview and has justification for believing (due to being told by the manager) that Jones will get the job. He also has justification for thinking Jones has 10 coins in his pocket because he saw Jones counting them. So, Smith forms the belief “the person who will get the job has 10 coins in their pocket”.
It turns out Smith does get the job and by luck happens to have 10 coins in his pocket.
So, Smith had a justified true belief.
Gettier has found cases where the tripartite criteria are present, yet knowledge is not.
So, Gettier problems are counterexamples to the joint sufficiency of the tripartite criteria.
It cannot be a correct theoretical definition of knowledge about what knowledge necessarily is if there are cases where the criteria obtain but knowledge does not.
Infallibilism is a variant of the tripartite view.
It claims that S knows p if and only if:
S believes p
p is true
S has infallible justification for believing p
This definition is meant to avoid the problem of epistemic luck by requiring that justification provide indubitable support for the truth of the belief.
In Gettier’s original counter-examples, Smith has justification for his true belief, but not infallible justification.
The manager could have been lying or deluded when telling Smith that jones would get the job. Smith could have been hallucinating when seeing Jones driving a Ford.
So the infallibilist definition is meant to avoid Gettier problems by explaining how Smith does not have knowledge.

Evaluation:

An issue for Infallibilism is that we know very little, perhaps nothing, infallibly.
We can never eliminate beyond doubt the possibility of some sceptical scenario like being a brain in a vat or an evil demon.
However, this criticism is weak because it does not attack the actual infallibilist definition of knowledge, only its usefulness.
Even if we couldn’t gain any knowledge infallibly, that wouldn’t mean the definition were incorrect. It would just mean that knowledge is unattainable.
It still seems theoretically coherent and without counter-example.
A stronger response which actually attacks the definition of infallibilism: it has stretched the concept of knowledge too far from its original social meaning.
The word ‘knowledge’ refers to an activity humans and perhaps some animals do.
‘Knowledge’ refers to whatever it is that enables us to do things like make planes fly and communicate over computers.
A definition where no knowledge is possible has failed to clarify this concept.
E.g. if ‘batchelor’ was defined in a way that failed to pick out any actual instances of batchelors, we would regard the definition as a failure.
So, we should say the same about knowledge.
If an infallibilist definition of knowledge fails to pick out any instances of knowledge. It fails to define the social construct called knowledge.
This suggests it has the wrong definition of knowledge.

No false lemmas

A false lemma is a false belief from which a conclusion is inferred.
Sometimes false lemmas can be used to infer true conclusions due to luck. This happened in Gettier’s original counter examples. Making no false lemmas a condition of knowledge is meant to avoid lucky true beliefs counting as knowledge.
E.g. Smith used the false belief that Jones would get the job to come to his conclusion that the person who will get the job has 10 coins in their pocket. The belief that jones owns a ford was also a false belief used to infer the conclusion that ‘either jones owns a ford or brown is in barcelona’.

Counter:

Barn facade case. Henry is in a county of fake barns. Every time he passes a barn, he thinks ‘there’s a barn’. Most of those beliefs are false, but there is one real barn.
Henry’s belief that ‘there’s a barn’ in the case of the one real barn is justified and true and isn’t inferred from any false belief/lemma. It is inferred from perception.
So, there’s no way for the NFL theory to say it’s not knowledge. Yet, it resulted from luck.
So, NFL cannot eliminate epistemic luck and thus fails as a definition of knowledge.

Evaluation

NFL could try to respond that Henry inferred his belief about the real barn from his general belief that he was in a county full of real barns, which was false since not all the barns were real.
However, this response fails if we simply specify the scenario with further detail.
it is logically possible for a person like Henry to be absentmindedly using perception to gain beliefs every time he sees a barn, without really thinking about being in a barn county or inferring beliefs about the current barn he’s perceiving from his previous perceptions of other barns.
So, the NFL definition cannot adequately cover all logically possible examples.

Reliabilism

Reliabilism is meant to solve Gettier cases and explain why lucky true beliefs are not knowledge, by removing the justification criterion.
The problem with justification is that the reason for the truth of a belief and the actual justification a person has can come apart. This opens a space for epistemic luck.
Smith relied on justification to reach his conclusion, either that the manager told him Jones would get the job, or seeing Jones driving a ford.
However the actual reason for the truth of the belief was that he got the job, or that Brown was in Barcelona
Reliabilism solves this by requiring that the truth of the belief be caused by the third criteria.
If Smith’s true belief were instead caused by a reliable cognitive process, then it couldn’t have been caused by luck. So the reliabilist definition is meant to avoid epistemic luck by causally relating the conditions so as to exclude luck as a cause of the true belief.
Reliabilism proposes that this can be solved by replacing the justification criterion with ‘reliably formed’.
Knowledge is reliably formed true belief. A reliable cognitive process would be something like memory or perception. A process which causes a high percentage of true beliefs.
This makes it a stronger theory than others because it explains how animals have knowledge.
Previous definitions of knowledge found it difficult to eliminate luck because luck could always appear as the cause of the true belief.
Reliabilism is in a stronger position than JTB and no false lemmas because it introduces the feature of a connection between the true belief and the other condition.
If the true belief was caused by a reliable cognitive process, then by definition it wasn’t caused by luck.

Counter:

The barn facade case shows that there can be circumstances where it is only by luck that a reliable cognitive process causes a true belief.
It was luck that Henry’s reliable cognitive process of perception caused a true belief in the case of the one real barn, because in the cases of the fake barns it caused false beliefs.
So it seems reliability actually can’t eliminate epistemic luck.

Evaluation:

Goldman solves this issue by adding a fourth condition, called ‘sensitivity’ or ‘ability to distinguish between relevant alternatives’.
Henry was not able to distinguish between the relevant alternative that he was in a fake barn county.
If he had been able to do that, he would not have formed the lucky true belief.
So, Goldman’s reliabilism can avoid epistemic luck and explain why Henry didn’t have knowledge.
It is also in a stronger position than no false lemmas because it can avoid the barn facade issue.

Conclusion:

The original tripartite view was shown to lack joint sufficiency by Gettier. Infallibilism was unable to salvage it, because it ended up eliminating knowledge rather than defining it.
No false lemmas seemed to be in a stronger position, yet could not address the fake barn scenario.
Reliabilism’s introduction of a causal relation between the criteria put it in a stronger position than the other definitions of knowledge. At first it seemed to face the same barn issue, but Goldman’s development of it allowed reliabilism to deal with that counter-example too, showing it to succeed where other theories failed.
So, the reliabilist definition of knowledge is successful