Recently I encountered Daniel Goldman a couple of times on Twitter.  The first time was during a discussion about the use of the words atheism, theism, agnosticism, etc.  The conversation was pleasant enough, especially when we consider how many of those kinds of discussions go!  The second encounter was a discussion about belief, and how it is defined.  Daniel disputed the definition of belief that I presented, and a discussion about said topic ensued.  There was a lot of back and forth that went on, which eventually led to him sharing an article he had written with me and me offering to look at it.  I also offered to share my thoughts on it with him.  Which is what this article is, the critique I offered.

Daniel’s article ‘I Believe That I Believe, or at Least I Think I Do’ can be found here:

And Daniel’s Twitter can be found here:

Daniel’s article is a short read, and I recommend reading that before reading this critique.  You should also give Daniel a follow on Twitter if you have a Twitter account.  Anyway, on with this article!

Definitions of Belief in Psychology

Daniel’s article begins with explaining his reason for researching the definition of belief.  He then states that while doing this research he could not find a reasonable definition of the word belief.  Here he is speaking strictly in terms of within psychology itself.  Mentioning at the end of the article that he may need to look within philosophy instead.  So, before continuing let’s look at some of the definitions offered in psychology.  This definition is from the Dictionary of Psychology (2002) from Ray Corsini:

(Corsini, 2002)

and this is the definition of doctrine that is used in the same dictionary of Psychology:

(Corsini, 2002)

Here is another definition that comes from the Oxford Dictionary of Psychology (2009-) from Andrew M. Coleman:

(Coleman, 2009)

and this is how the same dictionary of Psychology defines proposition:

(Coleman, 2009)

And here is another definition of belief from The Penguin Dictionary of Psychology 4th Edition (2009) from S. Reber, Rhiannon Allen, and Emily Sarah Reber:

(Reber et al, 2009)

Daniel states that he couldn’t find anything, and had been searching for a while.  It seems that there are dictionaries of Psychology out there with the definition in them.  Of course, it could be the case that I simply have an advantage over Daniel with regards to finding the information.  I have access to an on-line academic library, and perhaps Daniel doesn’t.  However, hopefully this might help Daniel in his search, or in his project.  Before moving on to the rest of Daniel’s article, let’s quickly return to his thought that he may have to look at Philosophy rather than Psychology for the definition of belief.

Definition of Belief in Philosophy

This is the definition given in the Stanford Encyclopedia of Philosophy (2021 Winter Edition) for belief:

(Schwitzgebel, 2021)

And this is from The Routledge Dictionary of Philosophy (2010) from Michael Proudfoot and Alan Robert Lacey:

(Proudfoot and Lacey, 2010)

And this is from A Dictionary of Philosophy (2002) from Antony Flew:

(Flew, 2002)

And this is from A Dictionary of Philosophy 3rd Edition from Simon Blackburn:

(Blackburn, 2015)

So, here we have several definitions of belief, from dictionaries that relate to both Psychology and Philosophy.  While there are variations in the definitions, there is a common thread.  That common thread is that a belief is to hold a proposition as true, or the acceptance of a proposition as true.  It is an epistemic attitude.  There is some disagreement as to how much evidence is required, or not required, as to whether it counts as a belief.  There is, however, lots of agreement that it is a propositional attitude, and it is about holding a proposition as true.

Kate Morgan and Belief

Though at the beginning of the article Daniel states that he couldn’t find anything on the definition of belief, he does admit a little further on that he did find some things discussing belief.  He first mentions an article he found from Kate Morgan on the psychology of a belief.  Daniel calls the article informative, but states that he found no definition of belief in the article itself.  It’s stated by Daniel that ‘[i[t’s just assumed that we know what a belief is’.  As I said previously in this article though, the definition from the Penguin Dictionary of Psychology states that in Psychology the term belief is ‘[g]enerally used in the standard dictionary sense’.  The article from Kate Morgan simply assumes that the reader knows what belief is, or has access to a dictionary to look the term up.  Something we see Daniel do later on in the article.

James Morgan and Belief

There is also mention of how it is used by James Morgan in a response to Kate Morgan.  The article quotes Morgan as saying that ‘a belief is just an opinion’ (Morgan in Goldman, 2019).  It also states that this definition is more or less in line with Duncan Riach’s discussion on belief and climate change, where he equates belief and faith (Riach in Goldman, 2019).  This definition is, quite rightly, questioned by Daniel.  As is asked by the article, is that really the case?  Are belief and faith identical?

Well, looking back over some of the definitions at the beginning of this article, some of them do define belief as something akin to faith, or mere opinion.  Some, however, also define it as something stronger than mere opinion.  Like I mentioned previously though, there does appear to be various opinions on what properties a belief has beyond it being a propositional attitude.  What conditions does the propositional attitude have that make it a belief?  Does it need to be an attitude made based on no evidence?  Does having evidence exclude it from belief?  Or can we have evidenced beliefs too?

Evidenced Beliefs

As Daniel mentions in his article, when we believe something we usually have some reason to believe it.  It can be based on ‘some kind of evidence, whether scientific or otherwise’ (Goldman, 2019).  Flew’s (2002) definition seems to agree with this also.  Flew (2002) defines belief as involving some degree of evidence, though not conclusive.  A belief is not as weak as a mere opinion, but also not as strong as knowledge (Flew, 2002).  It lies somewhere between being completely uninformed and being knowledge.  If, though, a belief is, at its core, a propositional attitude where we hold that proposition to be true, then a mere opinion would appear to be a belief too.  It would simply be a particular kind of belief.  If knowledge is a subset of belief because knowledge involves holding the proposition to be true, then why can mere opinion not be a subset of belief too if it involves holding the proposition to be true?

The Big Lebowski

If we think about the various definitions presented it seems, to me at least, that one necessary condition of something being a belief is that it is the acceptance of a proposition as true.  The question at hand, and present within the various definitions, is whether or not that is a sufficient condition for belief.  Are there other necessary conditions that must be present before it becomes sufficient for being a belief?  I would consider the acceptance of a proposition as true as being both a necessary and a sufficient condition for belief.  With other conditions determining the kind of belief it is, as well as relating to epistemic justification.

Daniel and the OED

We then find Daniel moving on to discuss belief as defined in the Oxford English Dictionary.  While Daniel lists all three definitions he found in the OED, he only criticises two of those definitions.  The two definitions being critiqued are ‘Something one accepts as true or real; a firmly held opinion’ and ‘A religious conviction’.  These are labelled (2) and (3) by Daniel in his article, so we will refer to them by those numbers here too for ease.  The first to be critiqued in the article is (2), so we will begin with that one.

OED Definition of Belief from Daniel’s Article

The focus of the critique of (2) is the use of the word opinion.  The critique can basically be summed up as such.  An opinion is a subjective stance.  While we can hold subjective stances as true, we can also hold objective statements to be true as well.  As objective statements are not opinions, then (2) is problematic.  Is that the case though?  Are objective statements distinct from opinion?

The OED and Opinion

The answer to those questions seems to be no, objective statements are not necessarily distinct from opinion.  Let’s look at the definition of ‘opinion’ in the Oxford English Dictionary (2015):

(OED, 2015)

As we can see, the OED defines opinion as ‘a view or judgement formed about something, not necessarily based on fact or knowledge’.  Daniel uses the sky being blue as an example of an objective statement in his article, so let us use that here.  If we consider the idea that the statement ‘the sky is blue’ is a judgement about the colour of the sky, then it seems to me that it is still an opinion.  It might be argued here that it is based on fact and knowledge, and therefore above opinion.  However, the OED states that an opinion is ‘not necessarily based on fact or knowledge’.  This means that while it doesn’t have to be based on fact or knowledge, it does not exclude an opinion being based on fact or knowledge.  This means that an opinion is not necessarily a subjective stance on something, it can be objectively based.  It must be admitted here that Daniel is using the word in the way that many people use the word though.  So, it is understandable why Daniel makes this argument.

Assenting to the Philosophical/Psychological Definition

Ted Lasso

It should also be noted here that Daniel somewhat assents to the idea that a belief is the acceptance of a proposition as true.  As well as assenting to the idea that a belief can be based on more than just opinion.  He considers the acceptance of an objective statement as true as a belief.  If he did not, then this criticism seems redundant.  If a belief is simply an acceptance of an opinion as true, then the fact that we can accept objective statements as true would simply mean that it is not considered a belief.  By disputing that (2) is sufficient because we can hold objective statements as true, Daniel is disputing the claim that beliefs are simply opinion based.  While also acknowledging that a belief is the acceptance of a proposition as true.

Misunderstanding the Dictionary

Definition (3) is up for critique next in his article.  To remind ourselves, (3) was simply ‘A religious conviction’.  Daniel begins with the question ‘Are all beliefs religious convictions?’ (Goldman, 2019).  This question shows a misunderstanding of the purpose of a dictionary.  The dictionary simply shows us how words are commonly used.  If the word belief is commonly used to describe ‘a religious convince’ then the dictionary will list that.  The dictionary is not attempting to give all the necessary conditions that makes something a belief.  It is not attempting to describe the ‘essence’ of belief.  So, when the dictionary describes belief as ‘a religious conviction’, it is not saying that a necessary condition of belief is that it is a religious conviction.  No, it is saying that one of the terms that English speaking societies describe ‘a religious conviction’ is the word belief.

His misunderstanding leads to other errors in reasoning.  Such as the question ‘are political beliefs religious convictions?’.  This question shows that Daniel is mistaking the descriptions of usage in the dictionary for necessary conditions.  Definitions (2) and (3) are not seen as separate descriptions, but instead as necessary parts working in conjunction to give us components of belief.  Daniel is looking at the definitions being given in the dictionary as some kind of Socratic ‘unitary’ definition, rather than in a Wittgensteinian ‘language game’ kind of way.  For which my only response to this is ‘what is a game?’.  Can Daniel give a definition of the word ‘game’ that contains all the necessary and sufficient conditions that covers all instances of ‘game’?  Or ‘art’?  Or even something like ‘cool’?

Daniel’s Definition of Belief

It is there that Daniel ends his critique of the definition of belief as commonly used.  From here he moves on to giving his own definition of belief.  The motivation here seems to be to give a definition that determines belief from other cognitive states.  The definition that Daniel gives for belief is ‘a learned and persistent cognitive state, which modulates behavior’ (Goldman, 2019).  He also gives us a couple of short examples for this definition.  The one I’ll focus on is the second example, where he discusses someone’s behaviour after watching a scary movie.

Daniel’s definition of belief from his article

The example describes someone that has watched a scary movie, and has problems sleeping later because they are worried about being killed by ‘some psycho nut job with an ax’ (Goldman, 2019).  Daniel states that this is not a belief.  While it does have the qualities of being a learned behaviour, and modulating behaviour, it is not persistent.  As it is not persistent then it cannot be a belief.  This definition is a very different definition to those that were presented at the beginning of this article.  While the definitions presented previously focus on the idea that a belief is an attitude towards a proposition, Daniel’s definition removes that element entirely.  Content, intentionality, and whether it is truth-apt is considered to be irrelevant in determining whether it is a belief.

The three necessary conditions for something to be considered a belief according to Daniel are it’s motivational power, whether it is a learned behaviour, and that the cognitive state is persistent.  Let us look at these individually to begin with, starting with the idea that the cognitive state is persistent.  What exactly is meant by persistent here?

This is how the word persistent is defined in the Oxford English Dictionary 3rd Edition (Stevenson, 2015):

(Stevenson, 2015)

Persistent Cognitive State

Definition (1) seems to be irrelevant to the definition offered by Daniel, so we will assume that he is using the word in the sense of (2).  The first question that’s brought to mind here then is, just how persistent does the cognitive state have to be for it to be considered a belief?  If we consider the example of the movie-goer.  Their cognitive state is continues to exist while they continue to worry about the axe murderer.  If that state lasts from when they go to bed to when the sun rises the next morning, is that considered persistent?  That seems to fit the bill of continuing to exist over a prolonged period.  Yet Daniel discounts that as an example of a persistent cognitive state.  What if that cognitive state returns the next night?  Is it considered persistent then?  Or does the break in the cognitive state during the day mean that it is not persistent?

Another question here is why must the cognitive state need to be persistent in order for it to be considered a belief?  Applying the principle of charity to Daniel’s use of persistent, let’s go with the idea that ‘continuing to exist over a prolonged period’ doesn’t necessarily mean ‘in constant existence’.  Instead let’s take it to mean a cognitive state that persists in an underlying sense.  So, the idea that we should look both ways when we cross the road is persistent because that cognitive state arises each time we cross the road.  Why is the ‘persistence’ of that cognitive state relevant to it being a belief?  What makes that relevant, rather than the fact that we accept it as true that we should look both ways when crossing a road?

Modulating Behaviour

Next let us look at the claim that the cognitive state must ‘modulate behaviour’ in order for it to be a belief.  By ‘modulate’ here I take Daniel to be stating that it must ‘exert a modifying or controlling influence on’ our behaviour.  It must be admitted here that beliefs can, and do, have motivational power.  Our belief that we should look both ways before crossing the road certainly has motivational power.  It motivates us to pause before stepping out into the road, and it motivates us to make sure the coast is clear before crossing.  Do they necessarily have to be motivational for them to be considered beliefs though?

According to the definitions displayed at the beginning of the article a belief does not have to motivational in order for it to be considered a belief.  While there are disputes about what makes something a belief in those definitions, it being motivational is not one of those disputed things.  We can think it is the case that our keys are in a bowl on the kitchen counter without it motivating us to go check if they are there, or motivating us to go get them.  That belief may motivate us when we leave the house and need to grab our keys.  It has no such motivational power until then though.  Under Daniel’s definition of belief, while we think it is true that our keys are in a bowl on the kitchen counter but not motivated by that it is not a belief.  It does not come to be a belief until we are motivated by it.

Now it could be argued here, of course, that there are times that it does motivate me, even if not all the time.  The fact that it does motivate me sometimes is enough to fit the motivational necessity.  So, let us try another example here.  I have a small collection of arcade machine replicas on a shelf in the room I stream from.  You can generally see them in the background when I do.  I think it is true that they are there.  Under the definitions listed at the beginning of the article, I believe they are on my shelf.  I am not motivated to do anything with them though.  I bought them purely for nostalgic and aesthetic purposes.  So, under Daniel’s definition I would not have a belief that they are on my shelf.  Which seems like an incoherent thing to say.  It becomes an example of Moore’s Paradox.  It would be the same thing as saying, ‘I know they are on my shelf, but I do not believe they are on my shelf’.


There is simply so much that could be said about the definition provided by Daniel.  We could discuss how Daniel’s definition means that beliefs are no longer about the truth of some proposition.  Which means that they become purely non-cognitive in nature.  So, we can no longer speak about the rationality of holding certain beliefs.  We can only speak of them in the sense of the kind of ‘every day’ rationality discussed by David Papineau.  It also means that beliefs now become separated from knowledge, or at least separated from ‘knowledge-that’.  Believing something to be true is no longer a thing according to Daniel’s definition, so they are no longer relevant to ‘knowledge-that’.

However, this article has gone on long enough.  Especially when you consider the length of Daniel’s original article.  I guess this article could be considered evidence of the ‘rule of 3’ when criticising an argument!  So, instead, I will conclude here.  Hopefully there is enough in this article to show that there seems to be no good reason to replace the definition of belief commonly used in psychology and philosophy.  There is, in fact, good reasons to reject Daniel’s definition as it does not describe how we think about belief.  It also breaks how we think about belief in relation to knowledge and epistemology.  Of course, Daniel will most likely provide counter-arguments for those presented here, and not only is that most welcome, but I look forward to reading any responses.


Blackburn, S. (2015) Oxford Dictionary of Philosophy 3rd Edition [Online], Oxford Reference. Available at https://www-oxfordreference-com/view/10.1093/acref/9780198735304.001.0001/acref-9780198735304 (Accessed 10th June 2022).

Corsini, R. (2001) The Dictionary of Psychology [Online], ProQuest Ebook Central. Available at (Accessed 10th June 2022).

Coleman, A.M. (2009) Oxford Dictionary of Psychology [Online], Oxford Reference. Available at https://www-oxfordreference-com./view/10.1093/acref/9780199534067.001.0001/acref-9780199534067 (Accessed 10th June 2022).

Flew, A. (2002) A Dictionary of Psychology [Online], Credo Reference. Available at https://search-credoreference-com/content/title/macdphil?tab=entries (Accessed 10th June 2022).

Goldman, D. (2019) ‘I Believe That I Believe, or at Last I Think I Do’, Medium, 15/10 [Blog]. Available at (Accessed 10th June 2022)

Proudfoot, M. and Lacey, A.R. (2009) The Routledge Dictionary of Philosophy [Online], ProQuest Ebook Central. Available at (Accessed 10th June 2022).

Reber, S. and Allen, R. and Reber, E.S. (2009) The Penguin Dictionary of Psychology [Online], Credo Reference. Available at https://search-credoreference-com./content/title/penguinpsyc?tab=entries (Accessed 10th June 2022).

Schwitzgebel, E. (2021) ‘Belief’, The Stanford Encyclopedia of Philosophy [Online]. Available at (Accessed 10th June 2022).

Stevenson, A. (ed) (2015) Oxford Dictionary of English 3rd Edition [Online], Oxford Reference. Available at https://www-oxfordreference-com/view/10.1093/acref/9780199571123.001.0001/acref-9780199571123 (Accessed 10th June 2022).

Other reading/viewing

Watch this video on YouTube.
Watch this video on YouTube.