Last Updated on Mon, 22 Jan 2024 | Critical Thinkers
Not only do actual statements of arguments typically include a lot of material that is inessential to the argument, they often exclude some of what is essential to the argument: some essential propositions are left implicit, when our task in reconstruction is to make the argument fully explicit. To say that a proposition is implicit in an argument is to say that it is part of the argument intended by the arguer - either as a premise or as the conclusion - but that it has not actually been stated by the arguer. To make the proposition explicit is simply to state it - in particular, to include it in our reconstruction of the argument. So a large part of argument reconstruction is to make explicit what was merely implicit in the original statement of the argument.
Consider a very simple case:
Does Mr Jenkins know how to read? Well of course. Didn't you know that he's a successful politician?
The arguer might be presumed to be arguing as follows:
P1) Mr Jenkins is a successful politician.
C) Mr Jenkins knows how to read.
So rendered, the argument is invalid. Nor would it be inductively forceful, because in order to know whether or not the premise makes the conclusion probable, you have to know a certain fact which the argument does not tell us - namely that successful politicians, at least usually, know how to read. But it is clear that, in drawing this inference, the arguer is making the assumption that successful politicians do know how to read. We have to make this assumption explicit:
P1) Mr Jenkins is a successful politician. P2) All successful politicians know how to read.
C) Mr Jenkins knows how to read.
We leave the reader to judge whether the argument is sound.
Let us now consider a more complicated and realistic example. The Moa was a very large, flightless bird which was native to New Zealand but which is now believed to be extinct. The Yeti is a large, white-furred ape which is probably only mythical, but which some people believe actually inhabits the Himalayas. Here is the argument.
The Moa is thought to have been extinct for at least 100 years. So naturally a great deal of scepticism is met by Paddy Freaney's claim to have seen a Moa in 1993. But as Freaney himself has pointed out, Freaney has climbed Mount Everest twice, and has not claimed to have seen a Yeti. So we ought to believe Mr Freaney's claim.1
The first task of argument reconstruction should be to identify, and if necessary to restate, the conclusion of the argument. Now in this case it might seem obvious that the conclusion of this argument is stated explicitly in the last sentence: We ought to believe Mr Freaney's claim. But this is not the most straightforward and informative way to put it.
1 Sunday Star-Times, Auckland, New Zealand, 10 December 1995, p. C5. This argument was actually given by Freaney himself.
Ask yourself: exactly what proposition is the ultimate concern of the arguer here? The arguer attempts to persuade us to accept a certain claim of Paddy Freaney's; but this does not really tell us the ultimate concern of the arguer until we specify what claim that is. Freaney's claim, the one that ultimately concerns the arguer, is the claim that he saw a Moa. So what the arguer is really trying to persuade us of is that Paddy Freaney did see a Moa. Here then is a case in which the conclusion is at least partly implicit; none of the sentences directly says that Paddy Freaney saw a Moa. Indeed, it might plausibly be suggested that the conclusion ultimately at issue here is one that is inferred from the proposition that Freaney saw a Moa, namely, that the Moa is not extinct. In that case it would be even more obvious that the conclusion of the argument is only implicit in the original statement of the argument. But let us not push the matter quite as far as that. For simplicity we shall take the conclusion to be that Paddy Freaney saw a Moa.
Once we have identified the conclusion, we must identify the argument's premises. What are they? It is clear that the first two sentences in the example are not premises of the argument. The arguer is arguing that Paddy Freaney saw a Moa; this proposition is certainly not supported by the fact that the Moa is thought to be extinct, nor by the fact that many people doubt Freaney's claim to have seen one. If anything, those facts make it less credible that Freaney saw a Moa. So, by the Principle of Charity, we should not take the arguer to be citing those facts in support of the conclusion that Freaney saw a Moa. Rather, the function of these propositions is perhaps to point out that resistance to Freaney's claim is unsurprising. (This sort of thing is very common: very frequently, an arguer will begin by explaining why there is opposition to his or her conclusion.) The only premise explicitly given is that Freaney has twice climbed Mount Everest, and has not claimed to see a Yeti. So our first shot at a reconstruction might look like this:
P1) Paddy Freaney climbed Mount Everest twice, and did not claim to see a Yeti.
C) Paddy Freaney saw a Moa.
Now this is a good start, but clearly it does not do justice to the arguer's intentions. As it stands, the argument is not deductively valid. Nor would it be inductively forceful if we were to write 'probably' before the conclusion. For suppose the only information you are given is that a certain person climbed Mount Everest twice, and that that person did not claim that he saw a Yeti. Does that give you any reason to infer that that person did see a Moa? Obviously not. PI does not itself make the conclusion probable.
Yet the argument as originally intended by the arguer surely does have at least some inductive force. Therefore the arguer must be implicitly relying upon some further premise or premises, which we should try to make explicit. One proposition which is surely relied upon by the argument, but which is not explicitly stated in the argument, is simply that Paddy Freaney claimed he saw a Moa. The second sentence almost states this explicitly, but does not quite do so. So we should include this in our reconstruction:
P1) Paddy Freaney climbed Mount Everest twice, and did not claim to see a Yeti.
P2) Paddy Freaney has claimed he saw a Moa.
C) Paddy Freaney saw a Moa.
This is better. We now have in the premises, it seems, a statement of all the particular facts about Paddy Freaney from which the arguer is inferring that Paddy Freaney saw a Moa. But still we have not made everything explicit. In order to see this, ask yourself, why does PI - in view of the fact that Paddy Freaney claimed he saw a Moa - support the conclusion that he did see a Moa? Well, it is something like this: The arguer seems to be responding to the charge that Freaney is a liar, that he wishes fraudulently to claim credit for a certain sort of zoological discovery. As against this, the arguer points out that Freaney climbed Everest twice without claiming a Yeti-sighting. But surely, reasons the arguer, if a person has twice climbed Mount Everest without claiming to see a Yeti, then he is not the sort of person to go around making false claims about seeing strange creatures (for such a person, having visited Everest, surely would not be able to resist claiming to have seen the notorious Yeti). Freaney climbed Everest twice without making such a claim, so he must not be such a person. If he is not such a person, then he didn't lie about the Moa. So, if he said he saw a Moa, he did see a Moa.
What is missing from the preceding reconstruction, then, is an assumed generalisation: roughly, that a person who has visited Mount Everest without claiming to see a Yeti is not a person who would fabricate the sighting of an exotic creature. This is a generalisation because it is not something said on the basis of knowledge about Paddy Freaney in particular. It is plausible to say, as a general fact about human nature, that a person who wishes to tell 'tall tales' about some particular kind of thing is unlikely to pass up a favourable opportunity to do so.
The argument implicitly depends upon this generalisation as a premise; hence in our reconstruction we should make it explicit:
P1) Paddy Freaney climbed Mount Everest twice, and did not claim to see a Yeti.
P2) Paddy Freaney has claimed he saw a Moa.
P3) A person who lies about sighting unusual or non-existent creatures would claim to see a Yeti, if he or she were to climb Mount Everest.
C) Paddy Freaney saw a Moa.
Note that, although we could have produced a valid argument by writing 'Everything Paddy Freaney says is true' (or an inductively forceful argument by writing 'Most of what Paddy Freaney says is true'), doing so would have failed to take into account the relevance of PI. If we had written that premise in place of the actual P3, then the conclusion could have been inferred from that premise along with P2, and PI would have played no role in the argument. This is important, because the arguer clearly means to give us a very specific reason to believe this particular claim of Freaney's; he or she is clearly not simply relying on Freaney's general honesty or trustworthiness.
This reconstruction of the argument now seems to include the relevant facts about Paddy Freaney upon which the arguer is relying, and it includes the generalisation upon which the arguer is (implicitly) relying. We could make its structure a bit more explicit by making an intermediate conclusion explicit (note that we have had to change the numbering of the premises):
P1) Paddy Freaney climbed Mount Everest twice, and did not claim to see a Yeti.
P2) A person who lies about sighting unusual or non-existent creatures would claim to see a Yeti, if he or she were to climb Mount Everest.
C1) Paddy Freaney does not lie about sighting unusual or nonexistent creatures.
P3) Paddy Freaney has claimed he saw a Moa.
C2) Paddy Freaney saw a Moa.
We will discuss the evaluation of arguments in more depth in the next chapter, but we might as well pause to evaluate this argument. The argument is not deductively valid. To see this, suppose that the premises P1-P3 of the argument are indeed true. In that case, Paddy Freaney was not lying, making up a story, in saying that he saw a Moa. But it does not follow that Paddy Freaney saw a Moa, for he could have been mistaken rather than lying. In that case, his claim to have seen one was not a lie, but nor was it true. The inference from CI and P3 to C2, then, is invalid. Therefore the truth of the premises would not rule out the conclusion's being false, so the argument as a whole is not deductively valid.
Might the argument be made inductively forceful, then? If the premises including P3 are true, then Freaney was not making up a story; but again, all we can conclude is that either he saw a Moa or he was mistaken. So whether the premises make it probable that Freaney saw a Moa depends upon which of these alternatives is more likely. For that, we would need more information. We would need to know the circumstances under which the alleged sighting took place - whether the light was good, at what distance it was seen, and so on. We would also need to know whether Paddy Freaney knows what the Moa looked like, hence whether he would know one if he saw one.
Since we simply do not know these things, we cannot estimate the inductive force of the argument very highly. All we can say with any confidence is that if the premises are true then either Paddy Freaney saw a Moa, or he was mistaken. We could rewrite the conclusion to state exactly that. There may be better arguments for the conclusion that Paddy Freaney saw a Moa, but given only the material at hand, the reconstruction we have given does seem to capture what the person giving the argument must have had in mind.
Let us now look at an example from Australia, that shows how making a generalisation explicit can reveal the weakness of an argument.
The suggestion that in order to protect children from sunburn, a rule should be instituted requiring all children at our school to wear a sunhat when they are outside after 11 a.m., is unacceptable. For clearly such a rule would be an infringement upon the freedom of the individual.
The conclusion, obviously, is that a sunhat rule at school would be unacceptable. The only explicit premise is that given in the second sentence. So we may begin with:
P1) A sunhat rule at school would infringe upon the freedom of the individual.
C) A sunhat rule should not be instituted.
Obviously, a generalisation is being implicitly assumed. A first shot at making it explicit, and which might well seem to capture what the arguer must have been thinking, might be this: that no rule which infringes upon the freedom of the individual is acceptable. So we write:
P1) A sunhat rule at school would infringe upon the freedom of the individual. P2) No rule which infringes upon the freedom of the individual is acceptable.
C) A sunhat rule at school is unacceptable.
Here is a deductively valid argument. If the premises are true, then certainly the conclusion is true. However, this does not establish that the conclusion is true. For that, we have to ask whether the premises are true - whether the argument is not only deductively valid but deductively sound.
Surely PI is true, in a sense: By definition, a rule of any kind restricts, hence infringes upon, the freedom of those to whom it applies - in this case the children of the school in question. So let us grant that PI is true. But look at P2. We often hear this sort of statement, and we are often so impressed by such a phrase as 'the freedom of the individual' that we accept the statement as true. We have the feeling that 'the freedom of the individual' is something important and valuable, therefore that anything which takes it away must be a bad thing. But, as stated here, this proposition is absurd. For as we just said, all rules 'infringe upon the freedom of the individual'. So what P2 amounts to is the absurd proposition that all rules are unacceptable. Unless you are a kind of radical anarchist, you have to conclude that this argument is not deductively sound.
Now there may be some less sweeping generalisation that the argument might employ instead of P2, which would be more plausible, yet sufficient to obtain the desired conclusion. But the arguer has not given any hint as to what generalisation this might be. So we cannot credit the
The practice of argument reconstruction | |
arguer with actually having supplied a helpful argument on the issue. They may have something more plausible in mind, but they have not conveyed it. This argument, by the way, is a good illustration of the importance of distinguishing between argument and rhetoric. It was by thinking carefully about the precise literal meanings of the words expressing the argument that we came to see that the argument is unsound. In order to do this, however, we had to have some courage: a phrase such as 'the freedom of the individual' is rhetorically powerful; nevertheless, it is just not literally true that every rule that 'infringes upon the freedom of the individual' is unacceptable; on the contrary, it is absurd to suggest it. |
Continue reading here: Covering generalisations
Was this article helpful?