link rel="shortcut icon" href="favicon.ico" Critical Thinking By Example Critical Thinking By Example
Critical Thinking By Example

 Chapter 9: Accepting Premises: The Question of Evidence
  Quiz 9.1 Quiz 9.2


Material covered in this chapter

  • 9.1 Premises based on common knowledge of the target audience.

  • 9.2 Premises based on experience

  • 9.3 Premises based on epistemic authority


We have seen one line of support for premises. In serial arguments a subconclusion works as a premise for a further conclusion, while the subconclusion itself has a subpremise set offered in support of it. It is clear that this cannot be a general model for accepting all premises. For if every premise itself requires a further premise in support, then either this process will go on forever, e.g., P1 requires P2 for support, and P2 requires P3 for support, and so on; or we will end up repeating ourselves (arguing in a circle). A good argument must have at least one “bottom level” premise—a premise that is not supported by further premises. Bottom level premises can be either main premises that are not supported by further premises, or subpremises that are not supported by further premises. Our question then is: When should we accept bottom level premises? We will examine three instances: common knowledge, experience and appeals to experts.

9.1 Premises Based on Common Knowledge of the Target Audience

For the most part, we have been thinking about evaluating the arguments of others. In thinking about when premises are acceptable it will help to turn the exercise around: let us think about constructing good arguments with acceptable bottom level premises. Consider these examples:

Examples 9.1-9.7: Candidate premises based on common knowledge.

9.1 The earth is moving thousands of miles per hour.

9.2 The earth is billions of years old.

9.3 Government officials should not steal public money.

9.4 A fetus deserves the same moral consideration as an adult person.

9.5 Marijuana should be legal.

9.6 The earth is the center of the universe.

9.7 There is a new pizza place in town.


Imagine the target audience is a typical university undergraduate class and you are asked to debate some topic. Premises 9.1-9.3 would be appropriate bottom level premises that would be accepted on the basis of common knowledge. Premises that are common knowledge do not require any additional support.

On the other hand, premises 9.4-9.7 will require some other form of support. There is significant disagreement about the moral status of the fetus, that is, its moral status is not a matter of common knowledge in the typical undergraduate class. The same is true of 9.5. Example 9.6 is an extraordinary claim that would take some very extraordinary evidence to convince people: you are not likely to find a single person in a university class who thinks 9.6 is true. Finally, 9.7 is probably not a matter of common knowledge: the fact that the restaurant is new probably means the existence of the pizza joint is not a matter of common knowledge.

The qualification that common knowledge is based on the target audience should not be overlooked. In societies with little contact with science and advanced technology, 9.1 and 9.2 may require additional support, that is, they could not be assumed to be common knowledge. On the other hand, these same cultures may accept 9.6. In a meeting of the amoralist club, 9.3 may not be common knowledge and require further support. Similarly, if you were giving an argument to the marijuana legalization club, you could safely assume that 9.5 was common knowledge. It would probably be safe to assume that 9.4 is common knowledge if you were presenting an argument at a Catholic Bishop’s convention.

It should be emphasized that just because some of these premises would not be accepted as common knowledge, does not mean that they could not be premises in an argument, and even a bottom level premise. 9.7, as we shall see, might be an acceptable bottom level premise if based on experience. A controversial premise like 9.4 or 9.5 will probably require additional arguments to be acceptable, so they are not likely to be accepted as bottom level premises in many contexts.

9.2 Premises Based on Experience

Evidence obtained from our sense organs can be the evidential support for a bottom level premise. The fact that someone has seen, heard, tasted, felt, or smelt X is often good evidence that X is true. Consider this argument: “Juan and Juanita said they saw a new pizza place in town. You know John loves pizza, we should take him there for lunch.” The acceptability of the premise, there is a new pizza place in town, is based on the experience of Juan and Juanita: they saw a new pizza place in town. Although it is not explicitly mentioned, it seems plausible that the premise, John loves pizza, is something you gained from experience: you have seen him enjoy pizza.

Of course not every premise that is said to be justified by experience is acceptable. Consider a similar issue: the question of creditability of eye-witness testimony often arises in courts of law. In both cases, the acceptability of what the experiencer reports will depend on a number of factors including:

·         the general reliability of the witness

·         the witness’ ability to experience what is being reported

·         how many witnesses there are

·         whether their reports are independent

·         the plausibility of the claim

The ‘general reliability of the witness’ refers to things like whether the witness is known to be honest and sober. As for the ‘ability to experience what is being reported’, the question here is whether the experiencer might plausibly be thought to have had the opportunity to experience what is reported. We should probably doubt someone’s experiential report that they identified (with unaided vision) a person five miles away in a large crowd. Similarly, someone who claims to have eaten the best pizza in the world has probably overstepped the bounds of what they can experience: it is pretty hard to try all the pizza in the world. In general, the more witnesses who see something, the more confidence we should have that the premise is acceptable. In the example above, the fact that both Juan and Juanita saw the same thing should give us more confidence that it is true, than if we only had the testimony of one or the other. Independent reports are ones where those experiencing something do so separately, and where the witnesses do not confer with one another about what they saw. If Juan and Juniata drove in different vehicles at different times and both report to you the existence of a new pizza place, you can be more confident that it is true than if they saw the place while driving together. 

Another general constraint on accepting premises based on experience is the plausibility of the claim being made. The more extraordinary a claim, the more extraordinary the evidence must be to support the claim. A new pizza restaurant in town is not something that unusual. Martians landing in the city square is unusual (to say the least). Juan’s testimony to the effect that he saw the new restaurant is sufficient in the former case; in the latter case it would probably be wise to hold off accepting Juan’s testimony about Martians until further evidence can be obtained.

9.3 Premises Based on Epistemic Authority

An epistemic authority is someone who has specialized knowledge about a particular area. (‘Epistemic’ means of, or related to, knowledge).  Many students are familiar with this concept from writing academic papers. If you are writing for your anthropology class, and you claim that humans have occupied North America for more than 40,000 years, you might cite an epistemic authority to support this claim. In practice, this usually would be a reference to a paper published in a reputable academic journal. Appeals to epistemic authorities do not happen only in academic settings. An argument made to your spouse that your house needs new wiring might be based on a premise that the existing wiring is a fire hazard. This premise, let us suppose, is based on a report from an electrician. Most of us are not competent to make such judgments. Electricians, with years of training and practical experience, are in a position to make such judgments: they are epistemic authorities on the matter.

 We identified some of the conditions for appealing to an epistemic authority in Chapter 6 in the fallacy of improper appeal to an epistemic authority above. Here are the conditions again:

·         The authority must be identified.

·         The authority must be respected in her field.

·         The matter must be one in which there is a reasonable consensus amongst the relevant experts.

·         The authority must be speaking about a matter in her field of expertise.

A premise that is accepted based on epistemic authority is only as good as the authority, hence the need to identify the authority. Identifying the authority makes it possible for your audience to check the authority’s credentials. Consider cases where the authority is not identified. You may rightly wonder whether you should believe the premise, since it is possible that the author simply made up the claim. Don’t take my word for it; there is good evidence to back up this claim. A recent study by a noted sociologist shows that 78.43% of all undergraduate essays simply make up epistemic authorities. You should not believe the appeal to an epistemic authority in the previous sentence because no authority has been identified. And in fact, I just made this statistic up out of thin air. If I had said, Professor Smith showed this in a study published in 2010 in the Journal of Better Sociology Statistics, then you could have checked my appeal to an epistemic authority. You would quickly find there is no such journal, and so have very good reason to doubt my appeal to an epistemic authority.

The condition that the authority must be respected in her field is perhaps obvious. If someone is not respected in their field it does not mean what they say is wrong. It does, however, point to the fact that we cannot accept a knowledge claim simply because they are in the field. It is possible, after all, for someone to get professional accreditation, to get a PhD or to become a licensed electrician, for example, and then lose his or her mind. On the other hand, the fact that someone is respected in their field shows that they have demonstrated their epistemic reliability to their colleagues.

The reason that there must be a reasonable consensus amongst the experts can be seen from the following example. You want to make an argument that appeals to the premise that abortion is immoral. You can find a number of experts that have claimed it immoral, but also a large number who deny that it is immoral. Choosing one expert over another is arbitrary in this instance. Rarely is there complete consensus in any field. You may find a few professors of biology that doubt that humans are descended from apelike ancestors, but the overwhelming majority of university biologists believe that we are descended from apes. There are a few climate scientists who deny that the earth is warming, or think that humans are not responsible for some of the global warming pattern. But the overwhelming majority of climate scientists believe that the earth is warming, and that humans are a major contributor to this pattern. (The popular press gives the few dissenters far more attention than the scientific community does).

A couple more comments about appeal to epistemic authority are warranted. The first is a matter of quantity: more is better (within reason, of course). Citing one author who concludes that humans are causing global warming is not as strong as citing three. At least as important is the quality of the citation: a statement made in a peer reviewed journal is more trustworthy, other things being equal, than popular press accounts of scientific research, or something overheard at a cocktail party. The peer review process ensures a greater likelihood that errors have been eliminated. Print books and journals tend also to be more reliable than most web-based material, since the former tend more often to be peer-reviewed than the latter. For example, there are many fine print books on critical thinking, and other things being equal, they would be more reliable than a website on critical thinking which is not peer reviewed. (Ahem).