How To Argue, Part 1:
How the Philosophers Get it Wrong
Introduction
The central purpose of arguing is to use arguments to settle a question for the time being, since that is generally the best that we can do. That means coming to a view about which of a set of rival propositions has stood up best to criticism and testing so far.
Philosophers typically contend that the best or right way to settle a question by arguing is to justify an answer to the question by deriving it from established propositions. On that view, a good argument is sound. That is, it is both valid and has true premises.
However, where people disagree on the answer to a question because they accept different sets of propositions, merely sticking to what one already takes oneself to know and drawing inferences is a recipe for a shouting match rather than a debate that leads to enlightenment. Instead of attempting to justify, by dogmatically deriving conclusions from already accepted propositions, a better way to proceed is to attempt to discover faults by assembling the rival answers to a question and assessing them by drawing from them logical implications that we can criticise and test. We should try, of course, to ensure that our arguments are valid; but the most useful of these arguments will not be sound. Rather, at best, they will indicate that an answer is false by deriving from it a logical implication that we take to be false (the falsity of the conclusion shows the falsity of a premise).
Arguing is also about using arguments to appraise rival answers to a question as better or worse, given the current state of the debate. Effective arguing involves the search for new answers to be tested, for new hypotheses (also to be tested) that logically imply one or other of the answers, and for new agreed observation-statements against which answers may be tested.
Justification versus Criticism
An argument is an abstract entity which represents a group of propositions, the premises, as logically implying one other proposition, the conclusion. For example:
All men are bald.
Socrates is a man.
Therefore,
Socrates is bald.
That argument is valid because it is necessary that, if all of the premises are true, then the conclusion is true (alternatively: it is impossible that all of the premises are true and the conclusion is false). Note that the first of the argument’s premises is false. This does not affect the argument’s validity, but it does make the argument unsound. A sound argument is a valid argument of which each premise is true.
In arguing, a person uses arguments to try to give or to get an answer to a question or a number of related questions. According to contemporary philosophers, a person argues well if
(i) he obtains an answer to a question by using a valid argument,
(ii) the conclusion of the argument is the answer to the question, and
(iii) the premises of the argument together justify or prove the conclusion.
From (iii) it follows that the premises are true and thus that the argument used is sound.
Unfortunately, we often do not know which propositions are true; and even when we think we know a proposition to be true, we are often mistaken. As a consequence, following the philosophers’ advice will often lead to an impasse. For example, the general propositions that Alf thinks are true logically imply that a certain conclusion is true but the general propositions that Betty thinks are true logically imply a different conclusion, one that conflicts with Alf’s. If Alf and Betty each offer valid arguments in justification of his or her answer, as the philosophers recommend, then each expounds his or her own view and denies the other person’s. But they seem unable to use arguments to try to discover which view is better. Each is locked into his or her own little mental prison.
Fortunately, we do not have to follow the philosophers’ advice. Instead of using valid arguments to try to justify his or her answer to the question, Alf and Betty may instead use valid arguments to criticise the rival answers. One way to do that is to use an answer as the premise in a valid argument of which the conclusion is a contradiction. If that can be done, Alf or Betty can learn that his or her view is mistaken.
Another way of criticising an answer is to use arguments to derive from it some propositions that may be tested against experience. We may then be able to conduct an experiment, or simply to make some observations, to discover whether the logical implications of the answer are consistent with what we observe.
It seems that the philosophers overlook a fairly obvious logical fact. They are well aware that, in a valid argument, the truth of the premises guarantees the truth of the conclusion; but they seem to overlook the correlative fact that, if an argument is valid, the falsity of the conclusion guarantees that at least one of the premises is false. Once we pay attention to that fact we can see clearly that valid arguments may be used in attempts to refute rather than in attempts to justify. We can see also that unsound arguments are valuable. If we are to argue successfully from the falsity of a conclusion to the falsity of at least one of the premises, the argument must be valid. But given that at least one of its premises is false, the argument is not sound.
This is a point of the first importance because the growth of our knowledge is typically revisionary. Galileo’s law of free fall was not just better than Aristotle’s, it contradicted it; Newton’s account similarly corrected Galileo’s; and Einstein’s account improved upon and contradicted Newton’s. Similarly, Kepler’s theory of the elliptical orbits of the planets improved upon and contradicted Copernicus’s account in terms of circular motions, and it was replaced and contradicted by Newton’s theory of approximately elliptical orbits, which was in turn replaced and contradicted by Einstein’s theory rotating approximate ellipses. The growth of our knowledge typically depends upon refutations, that is to say, the use of valid arguments to derive the falsity of a premise from the falsity of a conclusion. Good arguments must be valid; they need not be sound.
Agreement
We noted above that the philosophers’ idea of arguing, as justifying a conclusion using an argument with established premises, seems to be a barrier to arguers reaching agreement because each is blinkered by her own pre-existing beliefs. It is therefore surprising that philosophers generally contend that arguing well typically results in agreement between the parties to the debate, with all affirming, or all denying, or all doubting the same proposition. However, as in their commitment to justification, so in their commitment to agreement, the philosophers are taking a narrow view of arguing.
In scientific research, for instance, adherents of competing research programmes need to know how well their programme is doing in comparison with rival programmes. But they will often adhere to their research programme even if a rival is currently doing better, because they have ideas for further research that they think will turn around the situation. In a debate about the competing research programmes, the aim of the arguers is typically to obtain agreement on which answer is rated more highly than its rival(s) in the current state of the debate. They may obtain agreement about that, while those advocating one research programme affirm a particular answer and those advocating a different research programme doubt or deny that answer.
Sometimes the arguers do not even aim to reach agreement about the comparative rating of the available answers. A debate may be valuable if the advocates of rival propositions achieve a better understanding of the relative strengths and weaknesses of their own and other positions, even if they disagree with each other about what those strengths and weaknesses are. Each arguer might have aimed to engage in arguing with others to improve his or her own view of how the rival propositions stand in the current state of the debate without trying to convince others of that view.
Conjectures
A further problem is that contemporary philosophers seem to have a blind spot when it comes to the introduction into the debate of bold new conjectures which can be criticised by means of argument. We can note two roles that such conjectures may play.
First, a new conjecture may be proposed as a rival to the answers that are currently being debated. For example, at the start of the seventeenth century there was a debate about whether the heavenly bodies moved around the earth (the Ptolemaic view) or whether the earth and the planets moved around the sun (the Copernican view). The two views agreed that the heavenly bodies moved in circles with epicycles (small circles on large circles). Kepler’s contribution to the debate was to propose a bold new conjecture — that the earth and the planets had elliptical orbits around the sun. That conflicted with the entrenched view that the heavenly bodies moved in circular motions; but it was a much simpler theory than the others and gave a close fit to the observed motions, so it was generally accepted as the better theory.
Second, a new conjecture may be proposed to help to decide between the answers being debated. For instance, there is an ongoing debate over whether or not the area bombing of Germany from the summer of 1944 saved the lives of large numbers of Allied soldiers. A person arguing that it did may propound the following two conjectures:
(P1) area bombing hastened the end of the war by disrupting German industry;
(P2) area bombing hastened the end of the war by weakening German morale.
A person arguing that area bombing did not save the lives of large numbers of Allied soldiers may propose the following two conjectures:
(Q1) precision bombing would have been more effective than area bombing in disrupting industry;
(Q2) the destruction of civilian targets caused by area bombing made the Germans more determined to resist.
On the one side of the debate, conjectures (P1) and (P2), in conjunction with already accepted assumptions, logically imply that area bombing saved the lives of large numbers of Allied soldiers; on the other side, conjectures (Q1) and (Q2), in conjunction with already accepted assumptions, logically imply the opposite. The debate then moves on to testing those conjectures, that is, identifying logical implications of them that can be compared with observation-statements. If (P1) and (P2) survive testing but (Q1) and (Q2) do not, then the proposition that area bombing saved the lives of large numbers of Allied soldiers will be shown to be better than its negation in the current state of the debate; and vice versa.
Effective Arguing
We can draw a distinction between a narrow and a broad sense of arguing. In the narrow sense, ‘arguing’ is using arguments to get or give an answer to a question. We have just seen, however, that there are supplementary activities that can make arguing more effective by providing more material with which to argue. This new material contributes new propositions to help answer the question(s) to be settled, such as:
· making observations or performing experiments to provide observation-statements that can be used in refutations;
· propounding new conjectures that offer alternative answers to the question(s) being investigated, which new answers may be critically examined in their turn;
· proposing new conjectures that (in conjunction with accepted assumptions) logically imply one of the answers to the question(s) being investigated, which new conjectures may subjected to criticism and testing.
We may therefore use ‘arguing’ in a broader sense to include such supplementary activities; or we could use the term ‘debating’ to refer to arguing in that broader sense. In that broader sense, arguing (debating) encompasses the use of our imaginative problem-solving creativity in searching for solutions to a problem and also in searching for new ways to criticise proposed solutions.
