Chapter 1: Kinds and Sources of Claims
[last revised 5 September 2005]
© Craig DeLancey
We will understand a claim to be the kind of thing that is expressed by a sentence, and which is true or false. There are many kinds of utterances that do not express claims because they are not true or false. Examples include:
"I promise to be there at eight."
"Close the window, please."
Claims are instead expressed by sentences that are either true or false. Examples include:
"Three plus three is four."
"Water is formed of hydrogen and oxygen."
"One man alone shot J.F.K."
A simple test to distinguish claims is to ask if it makes sense to say the claim is true or is false. If someone says, "Close the window, please," it is inappropriate to say "False!" On the other hand, if someone says, "Three plus three is four," "false!" is the only reasonable response.
We will be concerned with only particular kinds of claims: determinate and objective claims. By determinate, we mean that a claim is true or false, and not both, and not neither. There might be some claims that are both true and false. For example, some people believe that a claim like "This sentence is false" is both true and false at the same time. It might be. There might be some claims that are sometimes not true and not false. But neither such cases will be our concern. We assume instead that there are many interesting and important claims that are determinate, and critical reasoning can help us decide whether we should believe these.
By objective, we mean that the claim could be expressed with a synonymous sentence that is universally true (that is, true everywhere at all times). Thus, "Today is Tuesday" might seem like a claim that is true sometimes and false others. But we will grant that it is objective because it is (assuming I am saying it at that time in that place) synonymous with the sentence "August 29, 2005 is a Tuesday in New York." Our interest is with objective claims because we can be secure that if they merit our belief they are likely to continue to merit our belief. This is all to deny that relativism is true of all claims. Relativism about some kinds of claims is the view that the truth value of those kinds of claims can change depending upon factors like the culture of the person stating the claim. Relativism might be true of some claims, but it is not true of all claims.
There are then some claims that are determinate and objective, and we will be concerned with these. This is just to maintain our common sense belief that "three plus three is four" is false regardless of who utters it; that our best science does tell us water is H20; and that either there were or there were not more than one shooter of JFK. From this point on, we will use the term "claim" to mean the determinate and objective propositions expressed by just such kinds of utterances.
In order for a claim to be determinate and objective, the relative utterance must be neither ambiguous nor vague. Ambiguous utterances are those that could have more than one determinate meaning. Examples include "Every year Dan tries to read Plato's Republic." This utterance is unclear: it seems to mean that Dan tries but is unable to read The Republic, perhaps because of its difficulty. But it could also mean that every year Dan tries to make time to read The Republic, presumably something he read in the past. Vague utterances instead are such that it is unclear when they are true. Examples include "Jones is pretty tall," or even "we are winning the war." Unless we take the time to define "pretty tall" or "winning the war" with clear criteria to test when they have been satisfied, it will be unclear when such an utterance is true. We combat ambiguity and vagueness by requiring that a claim be clarified. This is discussed most explicitly in chapter 5.
Critical reasoning is primarily concerned with helping you determine whether you should believe certain claims. Critical reasoning does this through providing a series of methods for evaluating certain kinds of claims, and also by providing knowledge about the nature of claims and how they often can be biased or inappropriate. We must begin this task with two questions: what kinds of claims are there? And, where do we learn about claims? We cannot hope to answer either question exhaustively, but we can answer them sufficiently for most problems of critical reasoning.
There are at least four kinds of claims:
2. Logical and mathematical
4. Methodological and metaphysical
It is important to describe each of these in turn.
Empirical claims are concerned with observable features of the natural world. Empirical claims have two essential features. First, to determine whether they are true, you must ultimately make some observations. I say "ultimately" because often empirical claims are the products of theories. Most scientists will believe that gravity acts the same on a planet orbiting the star Arcturus, if there were to be such a planet, as it does on Earth. They believe this not because they have been to that planet, but because they believe certain things about gravity, such as that all masses have a gravitational field. But this claim (that gravity will act the same on a distant planet) is still empirical for two reasons. First, it is the product of an empirical theory -- millions of observations of the nature of gravity have been made, giving observational corroboration to gravitational theory. Second, the claim could ultimately be checked. Someone might someday go to such a planet, and verify that gravity acts the same there.
This is a very important feature of empirical claims because, very often, empirical claims are made and treated as if they were not empirical, but somehow true for some logical reason or because they were obvious or because it would be good if they were true. Consider two politicians debating the effects of some economic policy. One claims reducing taxes on the wealthy will help the poor. The other denies this. This is an empirical question, but it will often be treated as something that follows from one's moral beliefs, and acrimonious debate will ensue.
A second essential feature of empirical claims is that they are what philosophers call "contingent." This means that they might be false. This is a difficult issue. Philosophers have done a great deal of work trying to get clear about possibility -- about how we should understand words like "might," "could," and "possibly." Such words are vague, in that there are kinds or degrees of possibility. However, in this case, we do not need to delve into these issues. Instead, we can take the weakest form of possibility as our guide. Philosophers sometimes call something "logically possible" if there is no contradiction in its denial. More specifically, for empirical claims, one can (in principle) consistently describe a world in which the claim is true, and another in which it is false. Here, to be consistent means to not contradict one's self -- to never both assert and deny some claim. In this sense, then, regarding any empirical claim, it is at least logically possible that it could be false or true. Thus, suppose the moon has a mass of 5 x 10^20 kilograms. This is an empirical claim. One could in principle describe a world in which the moon could have had a mass of 5 x 10^20 + 1 kilograms.
This may seem obvious but note that some claims are not contingent in this sense. 2+2=4, and 2+2 could not amount to anything else. It is not possible that 2+2=5. To say that 2+2=5 is possible would be to contradict yourself. For example, suppose someone does assert that 2+2 could be 5. Then we note that (2+2)-2 = 5-2, by their reasoning. (2+2)-2 is the same as 2+(2-2), which is the same as 2+0, which is 2. But 5-2 = 3. Hence it follows 2=3. But we know that 2 is not 3. So we have contradicted ourselves: we claim both that 2=3 and it is not the case that 2=3. The claim that 2+2=5 leads to a contradiction, and there is no way to keep the meanings of our mathematical terms and yet consistently describe an arithmetic in which 2+2=5.
Examples of empirical claims could include any of the following:
Jones is much more than six feet tall.
Water is formed of carbon and hydrogen atoms.
I believe that we should be kind to animals.
Jones said that Jimi Hendrix is the best guitarist.
The first claim is clearly one we can go and check, supposing we have Jones handy. The second might take some ingenuity, but we could ultimately prove that it is unlikely -- after all, electrolysis of water allows us to split it into hydrogen and oxygen, with no carbon left over. The third and fourth claims contain an ethical claim and an aesthetic claim. Ethical and aesthetic claims are not empirical claims. However, these are actually claims about what someone believes, and about what someone said. Presumably we can confirm these -- for example, by asking the speaker of the third claim what she believes, and by asking a witness whether Jones really made that claim.
Logical and mathematical claims are all claims that follow from definitions, assumptions, and deductive proofs alone. Deductive proofs will be described more thoroughly in chapter 3, but in brief these are methods that have, first, a finite number of statements that are explicit and agreed upon beforehand, and, second, a finite and clearly defined set of rules about how to make new statements from other statements. A statement that can be derived just using these other statements and consistent rules is will be (if we have developed our rules correctly) a valid conclusion. All of mathematics and logic aims for this rigor, although the rules and assumptions are often not explicit.
Some proofs, such as proofs of pure mathematics or logic, use deductive methods in which all of our assumptions are stipulations -- that is, these assumptions are not observations, but rather they are agreements about how we are going to use a term from that point on, or about certain relationships that we will assume hold between our terms. When we deduce a conclusion in this way, we call it a logical or mathematical claim. Examples might include:
Five is greater than three.
Three times twelve is thirtysix.
There is no largest whole number.
If a sentence (P) is true then the sentence (P or Q) is true for any sentence Q.
We could prove each of these claims by assuming things like what we mean by "six" and "times" and "whole number" and "or," along with some assumptions about how we reason from these things. We would not ever have to do some empirical investigation -- that is, we would not count out three beans to ensure these are fewer than five beans, nor buy three dozen donuts to ensure it comes to thirty six, nor count all the whole numbers to ensure there is no end to them, and so on.
The most important feature of logical and mathematical statements is that each must be either (1) necessarily true, or (2) recognized as an assumption, or (3) provable from explicit assumptions or necessarily true statements.
Things are much less clear when we consider value claims and methodological and metaphysical claims. Value claims are mostly claims about ethics or aesthetics. That is, claims that contain moral terms like "should," "good," "evil," "just" and their cognates; or "beautiful," "ugly," and their cognates. These kinds of claims are essential to our everyday lives, and perhaps getting clear about when we should believe them is the most important task philosophy can do. However, unfortunately there is little or no agreement upon universal standards for evaluating these value claims. We do know that they are neither empirical claims nor logical and mathematical claims. Without delving into profound philosophy, the best we can hope to do is get very clear about some important features of such claims. This is a task taken up in chapter 9.
The situation is equally difficult regarding methodological and metaphysical claims. Methodological claims are concerned with standards about how we accomplish certain tasks, but they often are not empirical or logical claims. Many of the claims made in this book will be methodological. For example, this is a true claim about science: it is necessary that a scientific hypothesis be testable. Note that this claim is not itself empirical or logical: it is not one we can confirm by checking the world or doing a deductive proof. Nor is it in any obvious way a value claim. Similarly, some philosophers have argued that our beliefs are best justified if they are the direct products of our sensory experiences (as opposed to, for example, our inferences) -- that is, concerned with things we see, hear, touch, and so on. But this view is not empirical, nor something arising directly from our sensory experiences. These kinds of claims are perhaps something we agree upon when we develop our method for seeking knowledge. Our method for justifying such claims will largely be beyond the scope of this book (taking us, as it does, from critical thinking into what philosophers call epistemology), although there is some discussion in appendix 1 on this topic.
Other examples of methodological claims include:
If a statement is scientific, then it is falsifiable.
Beliefs that are reliable are true.
All meaningful claims are either logical or empirical.
We also sometimes make claims about the fundamental nature of the universe. Philosophers call these "metaphysical claims." These are closely related to methodological claims because sometimes we assume things about the world in order to study it or form theories about it. For example, we might assume that every event has a cause, and this may in turn underlie a method that looks for causes to all events. Metaphysical claims are not empirical or logical and there is little agreement on how to evaluate them, but, like methodological claims, it appears that it is necessary to make some metaphysical assumptions. Examples of metaphysical claims include:
Every event has a cause.
There is a god that created the universe.
Simpler events are more likely to occur than complex events.
There may be other kinds of claims. Furthermore, some would divide these four kinds of claims further, distinguishing importantly different kinds of logical or empirical or metaphysical claims. For our purposes, however, this is unnecessary. Great strides up and out of the cave can be made if we keep in mind these four and their distinctions. We will be most concerned with empirical claims. These are the most practically significant claims. They also require a number of methods for evaluation.
Sources of Claims
The required methods for evaluating claims are as much dependent upon the sources of our claims as they are upon the kinds of claims in question.
Claims come to us in at least four kinds of ways:
2. Conclusions of Arguments
3. Products of Theories
Each of these sources needs to be evaluated in a distinct way. To see this, we need to get clear on what these sources are.
Observations are typically the best kind of evidence we have of whether a claim is true. This is especially the case for any empirical claim. I ask you if it is raining, and you look out the window and tell me. In general, observations should be uncontroversial. However, there are special concerns of which we need to be aware. Observations often require a bit of interpretation. And this means that subtle human biases can come into play. If the observer has some special interest in the results of her observations, for example, she may be inclined to interpret borderline observations in a favorable way. Also, many times we rely upon our memory of an observation, and not the observation itself. Any limitations on memory will therefore affect recalled observations.
Many observations are also used for generalizations: for reasoning to claims about a large group from observations of a portion of that group. This kind of reasoning requires special care if we are to ensure that some small number of observations is going to be representative of a larger group.
Arguments instead attempt to show us what must be true, given certain assumptions. Arguments alone can be used to show that some logical or mathematical claims are true, but otherwise, arguments cannot prove a claim but rather can show us that a claim must be true if we assume that certain other claims are true. This is no small accomplishment: we often overlook what follows from our beliefs. Furthermore, if some conclusion that follows from our beliefs is unacceptable, then we know that there must be something wrong with our beliefs. Logic is able to make absolutely clear how such evaluations can be made. Argument is therefore the most powerful tool of critical reasoning. Argument is also the most abused of critical reasoning tools.
Theories provide us with compact descriptions of the world. Their beauty and utility follows from their being able to predict things that have not yet been observed. But if we are to trust theories, we must have criteria to distinguish the better theories from the inferior ones. Our most successful theories are scientific theories. These theories must meet criteria of providing testable predictions that are corroborated. But this alone is not enough, since many competing theories can be corroborated. We need to rank these, and we do this by preferring those theories which have more predictive power than their rivals, are consistent with our other existing theories, and which are simpler. Understanding these criteria can help us recognize when a claim derived from a scientific theory merits our belief.
Most claims that we encounter, however, are parts of reports. If you watch the evening news, for example, you may be told that it is raining in Iowa City, or that scientists have observed a new particle, or that the war in Iraq is being lost. We don't observe these things, they are not the conclusions of arguments, and they are not the products of theories. They are rather merely claims that are offered to us by someone else. Evaluating these claims therefore requires us to evaluate the people or organizations that are their source.
Evaluating whether people or organizations are trustworthy is a task beyond the reach of traditional philosophy. Evaluating the sources of most reports is a messy affair, concerned with a range of contingent historical facts, requiring that we review such contemporary events as media and channel consolidation, and biases in privately owned media. And yet, this exercise is one we cannot shirk, since we will be unable to evaluate many, perhaps most, of the claims we encounter if we do not develop criteria for evaluating reports.