cje@topaz.RUTGERS.EDU (Ernst @ Sanctum Sanctorum) (12/02/85)
I'm looking for pointers to articles on the creation of a certain kind of
logic puzzle, specifically, the kind Martin Gardner calls
"Smith-Jones-Robinson" puzzles. These frequently involve a series of
propositions such as "The Englishman lives in the red house" or "Neither the
horse nor the dog is owned by the Norwegian". You are then asked to provide
the answer to a question like "Who owns the zebra?"
I know how to *SOLVE* them, but I'd like to know how to *CREATE* them. They
don't seem (to me) to be amenable to reverse engineering, plus there are a few
different types of these puzzles, presumably each with its own considerations.
I've seen puzzles in which people's names begin with A, B, C... and their
occupations start with a, b, c... and no one's occupation starts with the
same letter as his/her name. I've seen "positional" puzzles, in which
important clues are the relative positions of the various category members
(e.g., "The white house is on the immediate right of the green house"; "Milk
is drunk in the middle house"). I've seen puzzles in which you have males
and females and the clues run along the lines of "Neither Polly, the cat
owner, nor Mr. Smith have a green car", which tells you that Smith is male
(which wasn't known before) and that neither he nor Polly own the cat or the
green car.
It seems to me that these puzzles involve something more than standard truth
table formulae, that is, the propositions are not usually of the form "If
P then Q". So what form *ARE* they in? (I am not a mathematician or
logician, so if someone can tell me just how they *are* in truth table form,
I'd be interested.) For x people and y categories, how do you know the
minimum number of clues to provide to ensure the puzzle can be solved? How
do you know how often to mention any one category member (e.g., the white
house).
Again, I'm looking for pointers. Logic texts, articles, citiations in
_Mathematical_Reviews_, etc., are all appreciated. If the formulae are
relatively simple, and someone wants to mail them to me, I'd *really*
appreciate it.
Chris Jarocha-Ernst
ARPA: JAROCHA-ERNST@BLUE.RUTGERS.EDU
USENET: {inhp4!packard , seismo , allegra}!topaz!cjeags@pucc-h (Dave Seaman) (12/05/85)
In article <4253@topaz.RUTGERS.EDU> cje@topaz.RUTGERS.EDU (Ernst @ Sanctum Sanctorum) writes: >I've seen puzzles in which you have males >and females and the clues run along the lines of "Neither Polly, the cat >owner, nor Mr. Smith have a green car", which tells you that Smith is male >(which wasn't known before) and that neither he nor Polly own the cat or the >green car. The stated clue does not allow you to conclude that either Polly or Mr. Smith is a different person from the cat owner. Any puzzle whose solution depends on this deduction is defective. -- Dave Seaman {decvax|harpo|ihnp4|inuxc|seismo|ucbvax}!pur-ee!pucc-h!ags
kort@hounx.UUCP (B.KORT) (12/06/85)
If you would like to observe a master at work, pick up a couple of Raymond Smullyan's puzzle books. In particular, _Alice in Puzzleland_ and _What is the Name of This Book: The Riddle of Dracula and other Puzzles_. Although Professor Smullyan doesn't give a recipe per se, his Socratic style will lead the diligent reader toward a deep understanding of the use of puzzles as a teaching device. --Barry Kort
levy@ttrdc.UUCP (Daniel R. Levy) (12/07/85)
In article <2506@pucc-h>, ags@pucc-h (Dave Seaman) writes: >In article <4253@topaz.RUTGERS.EDU> cje@topaz.RUTGERS.EDU (Ernst @ Sanctum Sanctorum) writes: >>I've seen puzzles in which you have males >>and females and the clues run along the lines of "Neither Polly, the cat >>owner, nor Mr. Smith have a green car", which tells you that Smith is male >>(which wasn't known before) and that neither he nor Polly own the cat or the >>green car. > >The stated clue does not allow you to conclude that either Polly or Mr. Smith >is a different person from the cat owner. Any puzzle whose solution depends >on this deduction is defective. >-- >Dave Seaman {decvax|harpo|ihnp4|inuxc|seismo|ucbvax}!pur-ee!pucc-h!ags That's right; Polly could be the cat owner, and it would be true that she and the cat owner would not have a green car, of course. I think the exam- ple was inadvertently misstated. -- ------------------------------- Disclaimer: The views contained herein are | dan levy | yvel nad | my own and are not at all those of my em- | an engihacker @ | ployer or the administrator of any computer | at&t computer systems division | upon which I may hack. | skokie, illinois | -------------------------------- Path: ..!ihnp4!ttrdc!levy
kort@hounx.UUCP (B.KORT) (12/07/85)
For those of you who are looking for logic puzzles presented systematically, I refer you to the logic puzzle books of Raymond Smullyan. He is a Professor of Mathematics, Logic, and Philosophy at the University of Indiana. His books include _Alice in Puzzleland_, _What is the Name of This Book: The Riddle of Dracula and Other Puzzles_ and _To Mock a Mockingbird_. (That last book is an ornithological approach to combinatorial logic, the logic of detective mysteries and jigsaw puzzles.) Smullyan is one of the 6 greatest living mathematicians. He is enormously entertaining. He is also one of the greatest exponents of the Socratic Method that I have ever encountered. Many of his puzzles are presented in the form of dialogues (after Socrates and Lewis Carroll). Those of you who are also into Hofstadter will find the link between them in _The Mind's I_. When the world seems hopelessly irrational, vexing, and perplexing, it feels good to spend time with minds like Smullyan and Hofstadter. Enjoy. --Barry Kort
andrews@yale.ARPA (Thomas O. Andrews) (12/09/85)
In article <435@hounx.UUCP> kort@hounx.UUCP (B.KORT) writes: >Smullyan is one of the 6 greatest living mathematicians. He is >enormously entertaining. He is also one of the greatest exponents >of the Socratic Method that I have ever encountered. Many of his >puzzles are presented in the form of dialogues (after Socrates and >Lewis Carroll). Those of you who are also into Hofstadter will >find the link between them in _The Mind's I_. When the world seems >hopelessly irrational, vexing, and perplexing, it feels good to spend >time with minds like Smullyan and Hofstadter. Enjoy. > >--Barry Kort Has Smullyan actually done anything significant? Meaning no disrespect, but I think of Smullyan as clever version of Martin Gardner? Any support of the claim that he is one of the "6 greatest living mathematicians?" Greatest in what sense? He's a clever communicator,and a bright problem solver, but as a mathematician, what has he contributed to the field? Anyone out there with details? -- Thomas Andrews andrews-thomas@yale
hen@bu-cs.UUCP (Bill Henneman) (12/09/85)
He's not a mathematician, he's a logician.
ins_apmj@jhunix.UUCP (Patrick M Juola) (12/10/85)
In article <625@ttrdc.UUCP> levy@ttrdc.UUCP (Daniel R. Levy) writes: >In article <2506@pucc-h>, ags@pucc-h (Dave Seaman) writes: >>In article <4253@topaz.RUTGERS.EDU> cje@topaz.RUTGERS.EDU (Ernst @ Sanctum Sanctorum) writes: >>>I've seen puzzles in which you have males >>>and females and the clues run along the lines of "Neither Polly, the cat >>>owner, nor Mr. Smith have a green car", which tells you that Smith is male >>>(which wasn't known before) and that neither he nor Polly own the cat or the >>>green car. >> >>The stated clue does not allow you to conclude that either Polly or Mr. Smith >>is a different person from the cat owner. Any puzzle whose solution depends >>on this deduction is defective. > >That's right; I think the example was inadvertently misstated. The example was correct; there is a general (read: unstated) assumption in all commercial puzzles of this type that all individuals mentioned in a single clue are distinct. If this means that the puzzle is defective, so be it. Remember that most folks who solve these kinds of puzzles are cross- word puzzle freaks, not logicians. Pat Juola Hopkins Maths
kort@hounx.UUCP (B.KORT) (12/12/85)
Thomas Andrews writes: >In article <435@hounx.UUCP> kort@hounx.UUCP (B.KORT) writes: >>Smullyan is one of the 6 greatest living mathematicians. He is >>enormously entertaining. He is also one of the greatest exponents >>of the Socratic Method that I have ever encountered. Many of his >>puzzles are presented in the form of dialogues (after Socrates and >>Lewis Carroll). Those of you who are also into Hofstadter will >>find the link between them in _The Mind's I_. When the world seems >>hopelessly irrational, vexing, and perplexing, it feels good to spend >>time with minds like Smullyan and Hofstadter. Enjoy. >>--Barry Kort > Has Smullyan actually done anything significant? Meaning no disrespect, >but I think of Smullyan as clever version of Martin Gardner? Any support of >the claim that he is one of the "6 greatest living mathematicians?" Greatest >in what sense? He's a clever communicator,and a bright problem solver, but as >a mathematician, what has he contributed to the field? Anyone out there with >details? > >-- > Thomas Andrews > andrews-thomas@yale I should point out that the selection of "the six greatest living mathematicians" was done by a professional organization. It's been a while since I saw the (rather handsome) poster extolling the accomplishments of the 6 nominees. (I saw it on a bulletin board on the Stanford University Campus.) I regret that I don't recall the name of the organization or the other details. Perhaps someone on the net has seen the same poster and can provide the details. Smullyan has written a number of well-written books on logic and philosophy. His latest book is entitled _To Mock a Mockingbird_. It's an ornithological approach to combinatorial logic. It's a fun book, but I must confess that I didn't understand it at its deeper levels. Like Hofstadter, Smullyan is able to illustrate Godel's Theorem in a number of delightful ways. I like Smullyan because he makes mathematics fun. -- Barry Kort ...ihnp4!houxm!hounx!kort A door opens. You are entering another dementia. The dementia of the mind.
dsr@uvacs.UUCP (Dana S. Richards) (12/13/85)
> > Has Smullyan actually done anything significant? Meaning no disrespect, > but I think of Smullyan as clever version of Martin Gardner? Any support of > the claim that he is one of the "6 greatest living mathematicians?" Greatest > in what sense? He's a clever communicator,and a bright problem solver, but > as a mathematician, what has he contributed to the field? Anyone out there > with details? > I do not have them here but he has written two excellent texts/monographs on logic and I have found numerous journal articles by him but I do not yet have a comprehensive list. I am not a logician but I understand his contributions were considerable. I attended a lecture of his where the "light" portion was a (very) short proof of incomputability. Despite what another poster said I do believe that he is a mathematician; he is also a philosopher (with a strong bias, see Tao and 5000) which is natural for a logician I suppose. He mainly does puzzles and expositions now as far as I can see. Gardner is not clever? I see what you mean in that he (like Conway) interjects humor where ever possible. He is quite unlike Gardner in that he started out as a mathematician and drifted into exposition. Hence his books are really tough sledding for anyone without a math background. (Smullyan's interest in puzzles is longstanding; I believe he started doing retrograde chess problems in the 50's and was an early correspondent of Gardner in the early 60's.) I have all of his books in paperback, and am quite annoyed that Lady has not yet appeared in paperback. Hopefully Mockingbird will be in pb soon.
dsr@uvacs.UUCP (Dana S. Richards) (12/13/85)
> Remember that most folks who solve these kinds of puzzles are cross- > word puzzle freaks, not logicians. I would hate to characterize Smith-Jones-Robinson puzzle solvers as crossword freaks, or vice versa. Except for the kinship at some meta-level and the fact they both appear in Dell publications, I don't see any relationship. A true "freak" who is bored with the NYT xword will do cryptics, preferably variety cryptics (Atlantic, Harper's). I know some of you do these but everyone should try them. DO NOT do british cryptics unless you are both an anglophile and a masochist. Cryptics should be fair, so try those in Games magazine for starters. (Games also has the best american xwords also in my opinion (***'s). They also have logic problems but I never look at them.)
tmoody@sjuvax.UUCP (T. Moody) (12/15/85)
[] I believe that Raymond Smullyan developed the "tree method" for testing the consistency of sets of propositions in symbolic logic.