As the moderator of The Scholarly Kitchen’s comments section, this has been a tiring couple of weeks. Clearly Sci-Hub (which, it should be remembered has been around for 3 years) is driving a lot of passionate opinions among researchers, librarians, publishers and other stakeholders in the community. At times the sheer quantity can become overwhelming, and often there is more heat than light. It’s always fascinating to see how people interpret an article or interpret the comments of others.

One counterproductive tendency of internet arguing is the very common tendency to fall into the use of logical fallacies. As the You Are Not So Smart podcast notes,

If you have ever shared an opinion on the internet, you have probably been in an internet argument, and if you have been in enough internet arguments you have likely been called out for committing a logical fallacy…

The video below discusses some of the most common fallacies, and what I particularly like about it is that it makes it clear that employing a logical fallacy is often not a deliberate attempt to mislead, and that while it’s fine to point them out, one shouldn’t always assume a dishonest motive.

David Crotty

David Crotty

David Crotty is the Editorial Director, Journals Policy for Oxford University Press. He serves on the Board of Directors for the STM Association, the Society for Scholarly Publishing and CHOR, Inc. David received his PhD in Genetics from Columbia University and did developmental neuroscience research at Caltech before moving from the bench to publishing.

View All Posts by David Crotty


16 Thoughts on "Logical Fallacies and Internet Comments"

Yes…but…the concept of fallacy presupposes the concept of truth as anything more than opinion shared by many. This is different to the changeable concept of ‘truth’ held by scientists. Fallacies can be scientifically disproved only by reference to ‘truth’.

Relevant(?) quote from reddit that has achieved meme status:

“Q: If someone from the 1950s suddenly appeared today, what would be the most difficult thing to explain to them about life today?

A: I possess a device, in my pocket, that is capable of accessing the entirety of information known to man. I use it to look at pictures of cats and get in arguments with strangers.”

Of course that joke is also prone to start a flamewar here, as it has the word “access”. I am not participating in that though, just thinking the joke is funny, and true.

great timing SK! my favorite fallacy to spot is black & white. A: good. B: bad. got it?

i think it would also be useful to have a similar video on cognitive distortion. one complication is that a person has to be pretty self-aware to recognize their (our) own roles in these battles. i’m skeptical of people who don’t appear willing to accept new information that refutes their points of view. the best types of arguments are discussions with give-and-take, compromise, acknowledgement of nuance and complexity.

months of watching political debates have left me weary of ideologues.

Would there be any politics anywhere without recourse to fallacy? Would there be any religion? Would there be any education? Would there be any culture?

Sure, why not? I think you are giving the concept of fallacy way too much weight. False belief is not fallacy, as that term is used in logic.

There is an interesting paradox with the ad hominem fallacy. (Disclosure: I am a logician.) While it is indeed a fallacy in deductive logic, it can be sound reasoning in inductive logic, under the rubric “consider the source.” For example, if someone is trying to sell you something, or is clearly advocating a cause, it is reasonable to be suspicious of what they say. It is likely to be incomplete and may well be exaggerated, or even untrue. Thus who is speaking may be an important consideration in the reasoning.

Given that blog debates are frequently about the evidence for asserted claims,they are therefore often inductive in nature. Inductive logic is to a considerable degree the science of evidence. In such cases the claim that someone is committing an ad hominem fallacy may well be incorrect, even though they are pointing to the person, rather than to what they say. (I have not seen this paradox spelled out before.)

> if someone is trying to sell you something, or is clearly advocating a cause, it is reasonable to be suspicious of what they say.

It is always reasonable to be suspicious of what someone says. But this does not translate into evidence that what they say is false. This, for example, is why motive is not sufficient grounds for conviction in a criminal trial.

> it is likely to be incomplete and may well be exaggerated, or even untrue.

Indeed, it is questionable even that. I would question the basis for placing much weight on this likelihood. Sometimes, sure. But in the preponderance of cases? I would want to see evidence that this is the case. ‘Fact checking’ of politicians’ speeches, for example, show that while some have difficulty with the truth, others are truthful the vast percentage of the time.

> In such cases the claim that someone is committing an ad hominem fallacy may well be incorrect

A logician ought to know not to infer from a possibility to a fact. A sign or suggestion that something might be the case is not evidence that it is the case. The accusation that a person has committed a fallacy needs strong evidence in support of it.

I am not sure what you are saying, Stephen, since I am not referring to any specific case. It is the case that in criminal cases, motive is one of the three pillars of evidence, the other two being means and opportunity. I suppose the question is whether or not a reason to doubt what someone says is evidence against what they say. Probably not, or perhaps not up to a point, beyond which it is. (Never ask a philosopher a simple question.) The concept of evidence is one of the most difficult of all human concepts. Its complexity has sort of consumed my life, from time to time.

In any case I am merely pointing out that in many blog comment instances, where the charge of committing an ad hominem is laid, that charge is false. Your last sentence seems to concur, so I am somewhat confused by your response. But then I am easily confused. (Just ask my wife.) It is why I have made analyzing confusion my life’s work.

While I am at it I should mention a common fallacy that I have studied at length. I call it the Fallacy of the Mean. It occurs when the mean value of a statistical study is reported as though it were a true fact about the population being sampled. In reality statistical sampling theory tells us that it is extremely unlikely that the sample mean is accurate.

Here is a simple way to see it. In a properly done case we get a probability distribution for the true value. This gives us what are called confidence intervals. Science tends to focus on the 95% interval, where it is 95% likely that the true value is within the interval, which is centered on the sample mean.

But consider the 49% confidence interval. It is more likely than not that the true value is outside this interval, which is still centered on the sample mean. There is also the 40% confidence interval, the 30% interval, the 10% interval, etc. Each is centered on the sample mean, with growing confidence that the sample mean is simply false.

Yet it is the sample mean that is commonly reported as an established fact about the population in question. Moreover this value, which is known to be false, is often then used as an established fact in future studies.

How is this for a fallacy?

One man’s fact is another man’s fallacy. But both men might change their point of view. This change might be motivated by greater (or lesser) access to information and discovery. But it might also be motivated by changing patterns of prejudice or fashion. Any individual can also arrive at any point in between fallacy and fact (agnostic?), or at a point at which he accepts both as irrelevant equivalents. I therefore tend to see the ‘two’ as a continuum, and not necessarily a polar continuum, more a flexi-continuum on which we can and do play.

We have two different conversations going on here, which is fine but they need to be recognized. I am referring to logical fallacies, in the technical sense of specific, erroneous forms of reasoning. You seem to be using the term “fallacy” in a much broader, less technical sense. I do not understand your use of the term but that may be my fault.

Fallacies are becoming more popular as more people have online access. There are many good books explaining logical fallacies and enough resources online. I would recommend
They have a list with the most common ones.

Comments are closed.