In all honestly, I've always had a problem with the "recourse to authority" fallacy.

In theory it seems all sound and convincing: a claim should be judged purely and solely on the argumentation laid to support it but never on who made this claim.

However, in practice it simply doesn't seem to work this way...

  • For some reason websites like Wikipedia tend to have a strict No Original Research rule and demand all claims to be backed up with references to reputable (with an emphasis on "reputable") sources. This seems to be the exact opposite of what the reasoning behind the fallaciousness of recourse to authority would demand, since in Wikipedia a claim is judged purely and solely on the authority that made this claim and never on the argumentation to support it.
    • And yet, AFAIK, Wikipedia isn't being condemned for being the "poster child of the recourse to authority fallacy"; rather, what Wiki strives for is considered a Good Thing™.
  • In turn, claims made by Wikipedia (especially but not exclusively if they violate this 'NOR' rule) are not considered reputable and, IIUC, sourcing a claim in an academic paper with a reference to Wikipedia is a no-no.
    • Is science, therefore, guilty of this fallacy as well?

Actually, and ironically, scientists seem to themselves promote this fallacy more and more often nowadays! How many times have I heared already that 'there came time of experts', that people should stop believing in claims made by charlatans, and they especially should turn they backs on promoters of quack medicine and instead start trusting academic medicine…

It would seem to me that the fallacy of the 'recourse to authority fallacy' lies in it refusing to take into account one simple thing:

It is simply not possible to judge all claims oneself, for the following obvious reasons:

  • One may lack necessary competence to judge a claim.
    • While trying to support the NOR rule one of the Polish Wikipedians said:

Research means adopting a given hypothesis and then verifying it with this or that methodology - the analysis of sources is one of such methodologies. But you surely couldn't have meant this? If so, then as an author you need to at least have a PhD, because unless you are an ingenious self-taught man you are only able to do this on this level.

  • There are simply FAR too many claims for a single man's lifetime to judge!
  • Evidence needed to judge some claims may be unreachable by one man but reachable by another, for example due to classified material or privacy laws.

Am I misunderstanding the reasoning behind the fallaciousness of the 'recourse of authority fallacy'? What piece of its fallaciousness am I failing to grasp?

Final remarks:

  • I wanted to ask this question for a long time… I finally did because this exchange of opinions pushed me.
  • I admit this Q has a personal context… namely, I used to have my own opinion about everything and I used to challenge everyone, even reputable experts, who were holding a different opinion. Now I am strongly suspecting that I used to be (and, hopefully to a lesser degree, still am) an ignorant and arrogant man, that I was a prime example of the Duning-Kruger effect and that while I can and still should educate myself on the topics that interest me, until I get my PhD in all possible subjects I should also assume that people more knowledgeable and more experienced than me are, well, more knowledgeable and more experienced than me and probably their claims hold more water - even if I cannot understand them yet (this last part is important: my reasoning was that I used to reject claims backed up by arguments that didn't make sense to me but I was simply too incompetent to understand these (valid) arguments).
  • One of the intersting consequence of such considerations seem to be religious ones: if there is God, should we accept claims made by God before demanding God to prove them to us? However, I realize that this digression strays (way) too far.
  • 2
    Recourse to experts is simply unavoidable. ALL scientific research is based on the "method" of using already available results and evidence and go on... If every scientist has to re-check every fact/evidence/theory starting from the spherical shape of Earth to QM, we (humans) would have made no progress at all. Commented May 9, 2019 at 14:26
  • 1
    In every context : from daily life to scientific research projects we have to make a trade-off taking into account the limited amount of time and resources (brain ?) available. Commented May 9, 2019 at 14:27
  • 1
    I think the issue is that the fallacy is almost always intended to mean either appeal to an inappropriate authority (physicists that don't study climate change on climate change, mathematicians that don't study evolution on evolution etc.) or appeal to a literal authority (king, military, president). Refering to an expert in the field for the truthfulness of a claim is hardly considered invalid reasoning in society.
    – Cell
    Commented May 9, 2019 at 14:58
  • 1
    Appeal to authority is not always a fallacy, see Wikipedia, only appeal to nonauthority is. Alas, in practice it is often hard to tell the difference.
    – Conifold
    Commented May 9, 2019 at 21:07
  • "should we accept claims made by God before demanding God to prove them to us?" In Tibetan Buddhism they believe the Buddha to be omniscient. Nevertheless, they still are interested in the reasons the Buddha gave for his doctrine and not in simply following it because he said so.
    – Sam
    Commented Aug 1, 2021 at 13:57

6 Answers 6


You seem to be equating "peer reviewed publications" with "authority". This assumption is false, and your assumptions which hinge on that are then left backwards.

Anyone can get a paper published in a peer reviewed journal and have their work become a properly citable article. Even a child in first grade could do so if they possessed sufficient research and writing skills. All that matters is that the material is sound and correct.

The reason why first grade children do not publish papers is not because they fail to be an authority, but because they have not done the research and lack the writing skills. They lack that by quite a few years.

A less extreme example would be non-authoritative high school or college students. These people are writing acceptable research papers all the time, especially the college students.

Anecdotally, I have a paper published in a field for which I most certainly am not an authority figure at all and am far from being one. I just happened to study the topic well enough and participate in some research, and I got published. The topic has nothing to do with my career or with my college degree.

The citable research paper is not the authority. The citable research paper is the "argumentation used to support the claim" (to use your own words) which could be made by anyone, and the author is irrelevant. In fact, during the review process it is not uncommon for reviewers to be barred from knowing who the paper was written by.

Citable publications are often written by authority figures, but that is not what causes them to make quality citations. And authority figures have been known to make many statements which would be very poor citations as well.

  • (1) By your argumentation one has to read the paper before accepting what is being claimed. Not only the paper itself, but all papers it cites; and all papers the cited papers cite; and so on. Maybe even try to reproduce the result oneself. Otherwise, we're back to the beginning: the strength of the claim lies on the authority of the reviewers and the journal. However, to judge oneself if the paper indeed proves what it claims to prove, one needs considerable competences, so we're back to the beginning - and there were cases were papers published in reputable journals were proven wrong.
    – gaazkam
    Commented May 10, 2019 at 11:03
  • (2) is unfortunately too long to fit in a comment, so I'll present (2) as a follow-up question, today hopefully, tomorrow at latest.
    – gaazkam
    Commented May 10, 2019 at 11:04
  • 1
    @gaazkam One should read the paper before accepting what is being claimed. One should not simply see that there is a reputable citation provided and assume "Oh, there's a citation. Their point is made." We (yes, myself included) often do not check the contents of citation out of laziness or time constraints, but that does not mean that the citation by itself is sufficient evidence of anything. That is why a very short summary of the relevant point should be included. "According to Bob[1], 'The data for this event contradicted the Complex Model because...'" (1/2)
    – Aaron
    Commented May 10, 2019 at 13:49
  • 1
    @gaazkam (2/2) ... but neither should you check every word of the cited paper and every word of every paper it cites etc.. We do not reinvent the wheel every time we design a new car. We just accept that we believe the wheel works as described and build on top of it. If it is ever proven wheels do not work as we believe, at that time we will accept everything using wheels is also flawed. That's how it works. Build a tower supported from below. If the foundation breaks everything tumbles. If the foundation never breaks you're good. The foundation has broke before and many ideas fell with them.
    – Aaron
    Commented May 10, 2019 at 13:55

Your question is a good and important one. You are correct that mere appeal to authority is a bad idea, and that avoiding all possible appeal to authority is also not viable. Indeed the correct approach would be to not only attempt to gain a rough understanding of the concepts, evidence and explanations in the topic of interest, but also attempt to identify reliable sources. In other words, it should not be a matter of authority (conferred or perceived) but rather a matter of reliability as assessed by yourself. Naturally, the more knowledge you have in a particular field, the more accurate your assessment of reliability will be.

[...] I used to have my own opinion about everything and I used to challenge everyone, even reputable experts, who were holding a different opinion. Now I am strongly suspecting that I used to be (and, hopefully to a lesser degree, still am) an ignorant and arrogant man, that I was a prime example of the Duning-Kruger effect and that while I can and still should educate myself on the topics that interest me, until I get my PhD in all possible subjects I should also assume that people more knowledgeable and more experienced than me are, well, more knowledgeable and more experienced than me and probably their claims hold more water - even if I cannot understand them yet [...]

It is very commendable that you have become aware of the Dunning-Kruger effect and have assessed yourself to be potentially affected by it. Indeed almost everyone is susceptible to it, especially in fields outside their expertise. But there is in fact a sure way to escape it completely: You must learn basic classical first-order logic (preferably a Fitch-style deductive system and how to translate arguments from natural language into formal proofs). Once you do so, you can immediately identify logical gaps with no possibility for error except carelessness. You can also easily identify the assumptions that any argument relies on, and do the appropriate research to have a reasonably accurate guess of whether those assumptions are correct. This may lead you down an entire network of rabbit holes, and take a long time for you to assess just one claim, but in most cases relevant to daily life it can be done with a reasonable degree of confidence even if you are not an expert in the field.

For the same reason, you do not need a PhD in a field in order to adequately confirm or refute a claim in that area. For example, you do not need a medical degree to determine that homeopathy is almost surely nonsense and hence any website that promotes or mentions homeopathy without advising against it is highly unreliable. You also do not need any degree to identify most scams and frauds.

Also, it is still good to question everything (but without having your own opinion unless for very good reason), and request for justification of any part of what someone else says that you have doubts about. For example, if someone tells you that avocado can relieve arthritis, assist in weight loss and treat cancer, you should rightly be skeptical and assume by default that it is just a fad (you just need to dig a bit into the history of fads to recognize the same signals), and ask or search for solid evidence before changing your mind. Here, the good reason for the default opinion is that almost every health recommendation on the internet is unreliable and goes with the current fads, so for your own safety you should reject online health recommendations by default. But how to search? You can use your existing knowledge to narrow down to the most dubious parts of the claim, and focus on those. In this case, it would be the claim that avocados can treat cancer, and it is good to think of whether the cited studies are in fact good evidence for the claim. That is where basic science knowledge is very useful, because it gives you a rough idea of how the world works and tips you off to outlandish claims. For example, you should know that salt in high concentration kills almost anything in a petri dish.

One of the intersting consequence of such considerations seem to be religious ones: if there is God, should we accept claims made by God before demanding God to prove them to us? However, I realize that this digression strays (way) too far.

I also want to address a misconception ingrained into this aspect of your question. It is impossible that you actually have access to "claims made by God" unless you are claiming to have communicated directly with God. What is almost certainly the case is that some person X claimed that God claims Y, and their 'justification' to you is that some writing W claimed that God says Y. By basic logic, you can immediately notice that to even get to "God claims Y" you need to assume "If God says Y then God claims Y to be true." and "W is correct that God says Y.". I think it should be clear from this that such justification is on flimsy grounds, especially due to the second assumption, so you can focus your inquiry on that.

On the other hand, no matter how many incorrect belief systems exist, it does not imply that there is no correct belief system, nor does it imply that God does not exist. And if you do think that God exists, it is surely a great privilege if he proves anything to you by logically incontrovertible means! Usually though, questions about God's authority are related to questions about ethics, which are a quite different matter from factual claims.

  • " In other words, it should not be a matter of authority (conferred or perceived) but rather a matter of reliability as assessed by yourself " (emphasis mine). I like a lot this way of putting it. Can you suggest anything to read that elaborates on this distinction?
    – Sam
    Commented Aug 5, 2021 at 13:35
  • 1
    @Sam: I don't have suggestions since what I wrote is entirely my own. However, the key idea is that if you treat all sources equally you would just have a useless mash of conflicting opinions, but if instead you figure out which are more reliable in a given field then you can rely on them more for questions in that field. And this assessment of reliability should be done by yourself, mostly because that gives you the highest confidence in the assessment. If you lack time, it is possible to rely on someone else's assessment if he/she assesses many other sources very similarly to you.
    – user21820
    Commented Aug 5, 2021 at 13:46
  • 1
    As we can see in history and with current events, there are numerous 'authorities' (whether in some 'official' position or having some 'popularity among the masses') that are completely unreliable, but there is no way to conclude that without some chain of reliability assessments. Either we assess their reliability yourself, or we assess someone to be accurate (in our judgement) at assessing reliability, and rely on their assessment, or ... No matter what, we would have to do some initial work; we cannot just follow others blindly.
    – user21820
    Commented Aug 5, 2021 at 13:51

Douglas Walton and Marcin Koszowy note that the argument from authority may not always be fallacious: (page 1 in pdf)

This form of argument was traditionally categorized as an informal fallacy by the logic textbooks, but in recent years a revolution has taken place, and it is now regarded as a legitimate argument. It is nevertheless a dangerous one that can go wrong in some instances and be quite deceptive as a rhetorical tool for strategic maneuvering in argumentation.

To show how the fallaciousness of the argument arises they recall that verecundiam means modesty and John Locke's earlier description claimed one may feel it a "breach of modesty" to question authorities. Walton and Koszowy conclude: (page 3)

Anyone who backs his argument with the pronouncement of such an authority thinks the opinion cited ought to be final, and considers anyone who questions it to be impudent.

This explanation of why arguments from authority, especially the ones classified as arguments from expert opinion, can so easily and so often be fallacious. A fallacy can be defined as deceptive argument used as part of strategic maneuvering by means of which one party in argumentation is employing a clever tactic to get the best of his or her speech partner party unfairly.

They claim this deception may come from confusing two kinds of authority: cognitive authority and administrative authority.

The problem here is that the administrative type of appeal to authority typically seems like it should be less open to critical questioning than the epistemic type of appeal to authority. Therefore if there is some confusion about which category a given appeal to authority should fall into, it may be easy to treat an argument from expert opinion as though it were based on an administrative appeal to authority. Hence there is a normal tendency for the recipient of the argument to be overly intimidated by it, and to presume that it would be inappropriate to raise critical questions about it. So the fallacy in such a case resides in the reaction of the recipient to such an argument, but it may also arise from the way the proponent of the argument puts it forward. (page 5)

This may be enough to ground an answer to the title question:

How far does the fallaciousness of the recourse to authority fallacy reach?

Using expert opinion is not in itself fallacious in an argument. However, if one feels "it would be inappropriate to raise critical questions" then such an argument may be fallacious. The source of the fallaciousness may come either from the "reaction of the recipient to such an argument" or from "the way the proponent of the argument puts it forward".

John Locke, J. An Essay Concerning Human Understanding. Retrieved on May 11, 2019 from https://archive.org/details/in.ernet.dli.2015.218967/page/n599

Walton, D., & Koszowy, M. (2014). Two kinds of arguments from authority in the ad verecundiam fallacy. Two Kinds of Arguments from Authority in the Ad Verecundiam Fallacy. Retrieved on May 11, 2019 from Douglas Walton's website at https://www.dougwalton.ca/papers%20in%20pdf/14ISSAv08.pdf


How far does the fallaciousness of the recourse to authority fallacy reach?

The argument from authority reaches as far as the facts. Then it stops.

The official argument can be the basis of belief and action so long as there is nothing observed that says it is false. Then the authorized belief must be modified or abandoned. There are plenty of examples in science where orthodoxy and The Powers That Be have dictated beliefs which eventually had to be discarded because they fit so few of the facts.

As suggested by the question itself, recourse to authority is necessary because it is impossible for any given person to know everything. This situation requires a resort to sources more distant from the question under consideration.

In such cases the conclusion is expressed not in certainties, but in probabilities. A conclusion resting on an authorized argument might not be known to a certainty, but only to a likelihood of, say, seventy-five percent.

In that example, the argument from authority reaches three-quarters of the way to a hypothetical situation where the decision-maker has perfect information about the question under consideration. The authority fills in the gaps where the known facts leave off.


Source: Two Kinds of Arguments from Authority in the Ad Verecundiam Fallacy

1 - An argument appealing to an authority can never serve as a deductive inference, that is, it cannot be said that an affirmative argument is automatically correct just because a certain authority on the subject said that the same is true, that is:

• Ricardo Felício said that global warming does not exist.

• Ricardo Felício is an authority on meteorology.

• Therefore, Ricardo Felício is correct in his statement.

The logical structuring of the appeal-to-authority argument is incorrect, as it is fallacious, as it deduces that statement X, when spoken by B, is correct only because B is an authority on a matter related to statement X.

2 - The argument of appeal to authority can serve as an inductive inference argument, that is, you can use an authority to give credibility and reliability to your argument, and not an immediately correlational confirmation to it, making the premise valid, not true .


Yes, this has misunderstood the 'recourse to authority' issue. This isn't surprising; the issue is badly understood even among scientists an other authorities.

This issue goes all the way back to one of the fundamental worldview changes that began back in the Enlightenment: a shift from a reliance on personal authority to systemic authority. Modern forms of authority are ostensibly vested in systems and structures, not in people. To wit:

  • A police officer ostensibly has authority because he is sworn into a system of laws an rules an regulations, not because he personally has a gun and a badge
  • Government officials ostensibly have authority because it is granted to them by citizens, through a legitimate and structured process, not because they personally have or deserve to be in power
  • Academics and scientists ostensibly have authority because they adhere to rigorous system of evidence and reasoning, and don't depend on their own personal characteristics (like intelligence or insight) to make their claims

This emphasis on systems and structures eases the evaluative load on non-experts. One doesn't need to evaluate whether Person X 'knows his stuff'; one merely has to evaluate whether Person X is working within the standards and systems of this position to establish that his work is legitimate. Unfortunately, Enlightenment thinking hasn't penetrated society anywhere near as far as we like to imagine. Huge segments of humanity still view authority through the lens of individual puissance, as though power and virtue necessarily go hand in hand, an this leads to two problematic situations:

  • Personification of authority: those who assert that what person A says must be correct and valid — or at least intrinsically credible — because person A is sincere, forceful, and respectable. Thus some people continue to promote Flat Earth theories against the (systematically developed) scientific view because they assert that the people promoting the view are just as good and credible as the people opposing it.
  • Ad hominem reasoning: those who try to undercut systematically developed positions by attacking the personal characteristics or credibility of the person presenting the position. Thus some people try to discredit (systematically developed) climate science by asserting that climate scientists are just in it for the money.

Wikipedia's rules against original research are meant to block research of all sorts, because (in theory) Wikipedia is only meant to present established research, and (in practice) the web-forum design of the site is an extremely poor format for intellectual discussion and debate. In fact, it's such a poor format for intellectual debate that many topics on the site are dominated by factions and ideologues, which makes the site as a whole unsystematic, and untrustworthy for academic purposes.

  • An interesting answer, but I don't understand why should systems be inherently authoritative? Systems are made by people and may be hijacked. For example under coercive regimes universities and institutes usually cannot be trusted on matters related to the regime's official ideology, even if these matters are scientific in nature and even though under such regimes universities are still probably the closest thing to systemic science that is there. Likewise, if most scientists working on any given field were indeed 'just in it for the money' then their position would not be credible?
    – gaazkam
    Commented Jul 30, 2021 at 19:13
  • @gaazkam: Systems have a number of advantages: (1) They are comparatively objective, since they must be formally expressed in language; (2) they imply a fundamental consensus, since their objectivity makes them more subject to examination and critique; (3) they provide a framework for challenging and delegitimizing authority, which doesn't otherwise exist. Authoritarian regimes provide systems for controlling subjects while leaving the authorities immune to such systems; that's what makes them authoritarian. Commented Jul 30, 2021 at 19:44
  • @gaazkam: the nice part of systemic authority is that even people who want to be crapulent — police who just want to push people around, politicians on the grift, scientists in it just for the money — must still conform to the systemic rules that are in place. In many places in the world I could get out of a traffic ticket by slipping the officer a handful of tens; in Europe and the US I'd be more likely to get arrested, because officers don't want to risk their careers by violating the systems. Even crappy scientists have reputations to protect... Commented Jul 30, 2021 at 19:49

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .