Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site cca.UUCP Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!cca!g-rh From: g-rh@cca.UUCP (Richard Harter) Newsgroups: net.puzzle Subject: Re: How did they do in the test Message-ID: <6610@cca.UUCP> Date: Sun, 9-Mar-86 04:18:15 EST Article-I.D.: cca.6610 Posted: Sun Mar 9 04:18:15 1986 Date-Received: Sat, 15-Mar-86 20:52:20 EST References: <> <6490@cca.UUCP> <1432@mhuxt.UUCP> <> Reply-To: g-rh@cca.UUCP (Richard Harter) Organization: Computer Corp. of America, Cambridge Lines: 103 Summary: In article <> holmes@dalcs.UUCP (Ray Holmes) writes: > > ....... [Sonntag's comments deleted] ..... > > Not true see below. However the problem is flawed in another way: >Flootersnoot lied when he said that all three statements were necessary to >determine the ordering of the three students. Statement 2 is NOT necessary. >Consider that we have only statements 1 and 3 and that they are necessary. > >Consider statement 1: > If Wilson is the only one of the three that works at Adelphi then >this statement contains no information at all about the relative ordering >of the three students and is not necessary (also there is not a unique >solution). Thus Wilson plus one or both of the others works at Adelphi. > >CASE 1: All three work at Adelphi. > From #1, Wilson must be first, thus Jones and Smith must be second > and third. #3 then inplies that Smith finished ahead of the youngest > of the three. Thus the only possible ordering of the three is: > > Wilson 1st, Smith 2nd, and Jones 3rd. > >CASE 2: Wilson and one other work at Adelphi's. > Wilson must then be either first or second. If Wilson is second then > #3 gives no additional information and is not necessary (if fact > both orderings are possible). Thus Wilson must be first. Now the > reasoning of CASE 1 gives us the same ordering. > >Hence, in any case, we get the correct ordering without ever considering >statement #2. > Sorry, this is another logical trap, albeit a good deal more subtle. You are correct that case 1 is ruled out. Now the possible data cases are: A: Wilson and Smith work at the Adelphi book store. Wilson and Smith have red hair. Jones is the youngest of the three. B: Wilson and Jones work at the Adelphi book store. Wilson and Jones have red hair. Jones is the youngest of the three. Cases A and B are isomorphic, so we will consider case A only. The following table gives the only orderings consistent with each statement, considered singly: (1) Wilson, Smith, Jones Wilson, Jones, Smith Jones, Wilson, Smith (2) Wilson, Smith, Jones Wilson, Jones, Smith Smith, Wilson, Jones Smith, Jones, Wilson (3) Wilson, Smith, Jones Smith, Wilson, Jones Jones, Wilson, Smith Now consider, any two statements, taken as pairs. The following table gives the orderings consistent with each pair of statements: (1,2) Wilson, Smith, Jones Wilson, Jones, Smith (1,3) Wilson, Smith, Jones Jones, Wilson, Smith (2,3) Wilson, Smith, Jones Smith, Wilson, Jones That is, no pair of statements is sufficient to determine the order; all three statements are necessary. Where, then, is the fallacy in your argument? You say: "Wilson must then be either first or second. If Wilson is second then #3 gives no additional information and is not necessary (if fact both orderings are possible). Thus Wilson must be first. Now the reasoning of CASE 1 gives us the same ordering." Now it is true that if Wilson were known to be second then #3 would be unnecessary. However it is not known (from #1) whether Wilson is first or second. Given statement 1, #3 contributes the information that the order (Wilson, Jones, Smith) is not admissable. Your error lies in insisting that statement 3 must give information about each possible placement of Wilson. However all that is required of 3 is that it eliminate some of the possible orders implied by #1. What you are saying is equivalent to this: Premise 1: If I had the datum, 'Wilson is second', then statement three is not needed. Premise 2: Statement three is needed. Conclusion: Wilson is not second. However the correct conclusion is that you do not have the datum, 'Wilson is second', which is, in fact, the case. I hope this clarifies the matter. Tricky little bugger, ain't it. Richard Harter, SMDS Inc.