Tag Archives: Relations

Shells And Peanuts Again (And Again…And Again…In A Never-Ending GROUNDHOG DAY)

So one more time — but this time with feeling:  following Relevant Logic, we can avoid Classical Logic’s paradoxes of Material Implication, according to which the following statements are true…

1) If Cliff lives in Houston, Texas, then the earth has just one moon

2) If Cliff lives in Orange County, California, then Paris, Texas is the capital of France

…by insisting that the antecedent p be relevant to the consequent q.  The question of course now is:  what is the relation that makes p relevant to q?  In my previous post, one can, if they are sufficiently drunk, just barely make out the answer:  ‘whatever condition c along with (in the case of subjective probability) knowledge k makes the conditional probability of q equal to 1 given p is what makes p relevant to q.   Sometimes this ‘whatever’ is identical with an INFORMATION THAT relation (p is information that q); sometimes it is not.

( When the relation is identical with the INFORMATION THAT relation, c is the channel of information that allows p to be information that q. When the relation is not identical with the INFORMATION THAT relation, c consists in background conditions, especially causal laws, which, just as in the channel-of-information case, make the conditional probability of q given p 1. My current claim is that even when the relation is not identical with an INFORMATION THAT relation, it has a structure in common with the INFORMATION THAT relation.)

What I propose to do now in the next several posts is go through the> various examples I’ve mentioned previously (shell games, children with measles, wormy red apples, the ringing of defective doorbells, and so on) and a) work out when, in the example, the IF-THEN relation is identical with an INFORMATION-THAT relation and when it is not, and b) see what strange conclusions arise from this account of the relevance-making relation.  Maybe some of these will be so awful that one would prefer Classical Logic’s paradoxes of Material Implication.

In this post I propose to work through Dretske’s famous shell game example.  In that example, one will remember, a peanut is hidden under one of four shells.  I know from whatever reliable means that there is a peanut under 1 of the shells.  This knowledge reduces the probability that (a | the ) peanut is under shell #4 from 1 in whatever billions to just 1 in 4. Maybe my waffling here between ‘a’ and ‘the’ opens up a can of worms; I am unsure. I turn over shell #1.  There is no peanut under that shell.  The conditional probability that the peanut is under any given one of the remaining shells is now 1 in 3.  I turn over shell #2.  Empty.  The conditional probability that the peanut is under any given one of the remaining shells is now 1 in 2.  I say:

If shell #3 is empty, Then the peanut is located under shell #4

And what I say is surely true!  True, true, twue!!!!!  For if shell #3 turns out to be empty, then the conditional probability that the peanut is under shell #4 is 1.  The condition c that makes this conditional probability 1 given p is the characteristic that objects have — at least those objects large enough to be immune to whatever quantum weirdness — of persisting in one place unless molested.  The electron (at least according my remembered ((and almost certainly garbled in my memory)) pronouncement of a chemistry TA I had as an undergraduate) one finds orbiting this or that particular atom could have been on the nose of the Mona Lisa before getting observed, and might be there again a moment later.  But the peanut is not going to jump around like that, leaping to shell #1 one moment while unobserved, and onto the nose of the Mona Lisa the next moment.  It is going to stay placidly and inertially where it is — under shell #4 — while one turns over shell #3 and observes it to be empty.  Given this background fact about objects the size of peanuts, shell #3’s proving to be empty rules out the possibility that the peanut is not under shell #4.

Here the relevance-making factor — what makes the IF-THEN statement I uttered true — is also that factor that would make shell #3’s turning out to be empty INFORMATION THAT the peanut is located under shell #4.

But let’s turn back the clock.  I am now back at the point at which I am turning over shell #1.  Empty.  If I now jumped the gun and said (as if this were the movie GROUNDHOG DAY ((which I have not seen, by the way)), in which one atrocious day gets repeated again and again so that…”The phrase “Groundhog Day” has entered common use as a reference to an unpleasant situation that continually repeats, or seems to.”):

If shell #3 is empty, Then the peanut is located under shell #4

what I say would surely be false. Or at least it must be false if what I said in my first paragraph is true.  For were I to turn over shell #3 and discover it to be empty, the conditional probability that the peanut is located under shell #4 would not be 1, but 1/2.  So the same IF-THEN statement would be true at one time, and false at another.  And it would be true relative to my knowledge k at one time (I know that shells #1 and #2 are empty), and false relative to my lack of that same knowledge at a different time.

Not coincidentally, the (possible) emptiness of shell #3 being information that the peanut is located under shell #4 is something that is true at some times and false at other times, and is relative to one’s knowledge (or lack thereof) in exactly the same way.  In this particular case, what makes the If p Then q statement true is identical with what makes p information that q.

Now turn back the clock yet one more time (I warned you that this is another iteration of GROUNDHOG DAY).  This time I already know from a reliable source of information, even before I have turned over any shells, that the peanut is located under shell #4.  I turn over shell’s #1 and #2 as before.  Both are empty, as before.

But now, shell #3’s proving to be empty upon turning it over would no longer be INFORMATION THAT the peanut is located under shell #4.  This is so for at least two reasons.  First, according to Information Theory, “old information” is an oxymoron.  It is not information at all.  Shell #3’s turning out to be empty is not going to tell me, inform me, show me, that the peanut is under shell #4 because I already have this information.

Second, to generate information is to effect a reduction in possibilities.  In Dretske’s example of an employee selected by a succession of coin flippings to perform an unpleasant task, the eventual selection of Herman out of 8 possible choices reduced the number of possibilities from 8 to 1.  The selection of Herman generates INFORMATION THAT Herman was selected because of this reduction in possibilities.  But in my situation, already knowing that the peanut is located under shell #4, the number of possibilities regarding where the peanut is located is already just 1.  Turning over shell #3 to prove that it is empty does not reduce the number of possibilities from 2 to 1 — that number was 1 in the first place.  So in my situation shell #3’s proving to be empty does not generate, is not information that, the peanut is located under shell #4.

That the number of possibilities is in my situation just 1, not 2 means of course that the conditional probability that the peanut is located under shell #4 is not 1/2, but 1.  This means that shell #3’s proving to be empty does not make the conditional probability that the peanut is located under shell #4 equal to 1.  For that conditional probability was already equal to 1.  We are supposing that I already know that the peanut is located under shell #4, but I would not know this if the conditional probability were not already 1.  The very strange conditions that would have to obtain to make the conditional probability say, 1 in 2 would rule out this knowledge.  The peanut would have to exist under both shell #3 and shell #4 at the same time while unobserved, then “collapse” to a single location under one of the shells upon turning over the other shell and observing its empty condition.  So to say that I already know the location of the shell is to say that the conditional probability the peanut is at that location equals 1.

Now in the first paragraph of this screed I said (maybe ‘pontificated’ is the better word):

…whatever condition c along with (in the case of subjective probability) knowledge k makes the conditional probability of q equal to 1 given p is what makes p relevant to q.

Here my knowledge k (the peanut is located under shell #4) presupposes certain conditions c (the peanut does not exist in a kind of locational smear when unobserved, only to collapse to a single location when an observation is made).  Here p (shell #3 proves to be empty) along with k and the presupposed c definitely does not make the conditional probability of q equal to 1.  This conditional probability was, given k and its presupposed c, already 1.  So in my situation, p is not relevant to q.

So were I, in my situation of already knowing that the peanut is located under shell #4, to  utter GROUNDHOG-DAY-wise:

If shell #3 is empty, Then the peanut is located under shell #4

My statement would be false for exactly the same reason that the following is false:

If Cliff lives in Houston, Texas, then the earth has just one moon

In both cases, the antecedent is irrelevant to the consequent by failing to make the conditional probability of the consequent 1, rendering the corresponding IF-THEN statement false.  The antecedent “If shell #3 is empty” is in my situation irrelevant to the consequent “the peanut is located under shell #4” in exactly the same way that “Cliff lives in Houston” is irrelevant to “the earth has just one moon.” (In exactly the same way?  Yes, at least according to the perhaps narrow definition of relevance I postulated above.  But does this narrowness weaken my claim?  Might the emptiness of shell #3 be relevant to the peanut’s being located under shell #4 in some ((perhaps)) vague way even given my knowledge k?)

To re-iterate (this is a GROUNDHOG DAY post after all), the shell statement is false in my situation for exactly the same reason that “shell #3 is empty” fails to be information that “the peanut is located under shell #4.”  In this particular case, the relevance-making condition which is lacking is identical with an INFORMATION THAT relation.

If so, however, one is faced with a consequence that may strike some as at least equally unappealing as the paradoxes of Material Implication.  (Warning:  I am about to wallow in more GROUNDHOG DAY iterations.)  For when I utter:

If shell #3 is empty, Then the peanut is located under shell #4

the statement I utter is false, but when you hear:

If shell #3 is empty, Then the peanut is located under shell #4

and your situation is such that you have seen both shells #1 and #2 are empty and you do not know that the peanut is located under shell #4, the statement you hear is true!  The same statement is both true and false at the same time, given different situations.  Put another way, what is true or false (at least for a certain class of IF-THEN statements) is not the statement, but the statement as it shows up in a particular situation.

At least in the case of subjective probability, then, truth is relative in much the same way that Galilean motion is relative.

On a purely autobiographical note, I am not sure this relativity bothers me any more than Galilean relativity (there is the possibility of an ultimate reference frame) or for that matter Einsteinian relativity (there is no ultimate reference frame which would assign a single value to the speed of a moving object) does.  The idea that a person walking inside a flying jet is moving at a speed of 1 mile per hour relative to the reference frame of the jet but at a speed of 501 miles per hour relative to the reference frame of the earth (suppose the jet’s speed is 500 miles per hour) is perfectly intuitive even though it means a contradiction is true (the person is both moving at a speed of 1 mile per hour and is not moving at a speed of 1 mile per hour).

If shell #3 is empty, Then the peanut is located under shell #4

Is both true and false at the same time seems to me to be intuitive if one casts it as a matter in which a conclusion’s following (not following) from its premise hinges upon what other knowledge or evidence one has (does not have).  But I do suspect that some would prefer to this relativity of truth and the attendant tolerance of contradiction the weirdness of Classical Logic’s Material Implication which arises from treating Implication as purely truth functional.

If shell #3 is empty, Then the peanut is located under shell #4

is variously true or false — even at the same time — depending upon the already-existing knowledge (or lack of it) of the person uttering or hearing the statement.  By contrast, the following statement is true regardless of what anyone knows, and true in any situation:

If the peanut is located under shell #4, Then the peanut is located under shell #4

In other words:

If p Then p

That the peanut is located under shell #4 clearly suffices to make the conditional probability that the peanut is located under shell #4 1.  So according to my account of what makes p relevant to q, p is relevant to p. p is relevant to itself.  p is in a relation to itself.  I am of course beginning to sound very weird (or maybe weirder) and very Hegelian…and I am beginning to wonder if I can get out of this weirdness by talking about 1-place relations, which are perfectly respectable mathematically.  (Not just 1-place relations!  0-place relations are also quite respectable mathematically!  What is more, Chris Date’s Relational Algebra recognizes two 0-place relations, TABLE DEE which is identical with the that weird proposition in logic TRUE, and TABLE DUM, which is identical with the equally weird proposition in logic FALSE!!!!!!!)

In this section of my post, I will decide that I am Relational-Algebra-weird by treating “If p Then p” as a 1-place INFORMATION THAT relation.  This in turn is part of my larger project to go through each example of IF-THEN statements I’ve adduced in previous posts and decide whether the relevance-making RELATION is in that particular case an INFORMATION-THAT relation or not.

Remember that to generate information is to reduce the number of possibilities to one.  When Herman is selected through 3 successive coin flips out of 8 candidates to perform the unpleasant task, the number of possibilities is reduced from 8 to 1.  The probability of Herman’s getting selected was initially 1 in 8, then became 1.  Whenever any event occurs, some states of affairs comes to obtain, some thing acquiring some property, the probability of that occurrence goes from 1 in (some usually gargantuan number) to just 1.  So any occurrence of p (Herman’s getting selected, shell #3 proving to be empty, a ruby having formed through whatever geological processes exactly one mile underneath where I happen to be sitting now typing this disreputable screed into a WordPress blog, the doorbell’s ringing) generates information.  Sometimes the occurrence of p generates information that q (that the peanut is under shell #4…that someone or something is depressing the button outside….).  But whatever else the occurrence of p generates information about, it generates at the very least the information that p.  Herman’s selection generated the information that Herman was selected, whether or not this information gets transmitted from the source situation in which the selection occurred (the room where the employees performed 3 coin flips) to the situation which is waiting for the information (the room where the boss is sitting).  When the information does get transmitted from source to receiver, the INFORMATION THAT relation is a 2-place relation comprising two situations, source and receiver.  When the information does not get transmitted, but stays where it is in the source, the INFORMATION THAT relation is a 1-place relation, comprising simply the source situation.

When the relevance-making relation that makes If p Then q true is an INFORMATION THAT relation, the occurrence (obtaining, existence) of p generates the information that q.  We have just seen that the occurrence (obtaining, existence) of p generates the information that p. So we get:

If p Then p

as a 1-place INFORMATION THAT relation.  Rather than saying, rather weirdly and rather Hegelianishly, that p is related to itself by virtue of being relevant to itself, we simply say that there exists a 1-place relation comprising the source at which the information that p was generated, and only that source.  This remains an INFORMATION THAT relation even though nothing ever tells me, informs me, shows me that, for example, a ruby exists exactly 1 mile beneath where I am now sitting, typing this disreputable screed into WordPress, or that the peanut is in fact underneath shell #4.  It is just a 1-place, not a 2-place relation, and an INFORMATION THAT relation to boot.

So in all of the following,

If a ruby exists exactly 1 mile underneath where I am now sitting, Then a ruby exists exactly 1 mile underneath where I am now sitting

If the peanut is located underneath shell #4, Then the peanut is located underneath shell #4

If Herman was selected to perform the unpleasant task, Then Herman was selected to perform the unpleasant task

the general relevance-making relation, i.e., the occurrence (obtaining, existence) of p making the conditional probability that p equal to 1, is identical with an INFORMATION THAT relation.  (My ((probably non-existent)) reader will remember that the relevance-making relation is not always an INFORMATION THAT relation.)

And this (after having brought in a ruby example and a Herman’s getting selected example) concludes my working through of most of the peanut-under-a-shell examples.  I still have one more peanut and shell example to work through, namely,

If I turn over shell #4, I will see the peanut

which I will work through in a future post.

Today’s homage to Plato’s SYMPOSIUM is Channing Tatum, who has recently appeared in MAGIC MIKE II.

Channing Tatum is the very walking, talking, breathing, living definition of the words ‘age 35 and beautiful and sexy.’  One of these days I will get around to contemplating Plato’s Form of Beauty itself.  For now, though, I will rest content just contemplating the form of Channing Tatum.

July 18, 2015:  extensive revisions made in probably futile attempt to hide the vastness of the extent of my confusion.

July 21, 2015:  made one more revision in order to try to hide the lack of control I have over the subject matter.

August 02, 2015:  made yet another revision for the same dubious reasons as listed above.

This is a version of the post below, revised so as to try to eliminate a number of confusions.

The Wikipedia article Argument (linguistics) starts its discussion of the argument/adjunct distinction by asserting that an argument is what is demanded by a predicate to complete its meaning, while an adjunct is not so demanded.  For example, if someone asks me “What is Joe eating?” my answer would be drastically incomplete if I replied “eats.”  My answer would still be drastically incomplete if I supplied just one argument, ‘Joe’, to say ‘Joe eats.’  Only when I supply a second argument, say, ‘a fried egg’, would my reply not create a sense of a question ludicrously left hanging and an answer simply not given.  The predicate _eats_ has two parameters ( shown here as ‘_’) demanding two arguments, such as  ‘Joe’ and ‘a fried egg’ for my reply to make any sense.

( This example, of course, is my own; I am offering it (maybe tendentiously?) in order to make drawing certain conclusions more natural. )

‘[I]n the kitchen’, however, is an adjunct, since nothing would be left ludicrously left hanging in the air were I to leave that phrase out of the proposition “Joe eats a fried egg in the kitchen.”  The predicate eats does not have a parameter demanding something like ‘in the kitchen’ as an argument.

This criterion — i.e., what is demanded by a predicate to complete its meaning … henceforth I will call this the ‘demands criterion’ — runs into trouble when one notices that sometimes eats demands two arguments, but sometimes demands just one.  One might say:  “Joe goes into the kitchen.  Joe is ravenous.  Joe sees food.  Joe eats.”  ( Imagine a novelist or short-story writer working in a certain style.)  The argument ‘a fried egg’ is not demanded in this particular piece of discourse.

But if ‘a fried egg’ is an argument, not an adjunct to eats, it would seem one would  have to abandon the ‘what is demanded by a predicate to complete its meaning’ criterion and find another criterion for what is to count as an argument and what is to count as an adjunct.  This a contributor (doubtlessly not the same person who put forward the ‘demands’ criterion) to the Wikipedia article cited above tries to do.

But if one wants to retain the demands criterion, they (I am intentionally using ‘they’ as a genderless singular pronoun) can assert that two different predicates, each with a different number of parameters, may get invoked when someone utters  ‘eats’ in a stretch of discourse.  Sometimes the one-place predicate _ eats is invoked, sometimes the two-place predicate _eats_.   Which predicate one uses is optional, depending upon what they feel is called for by the situation and what they want to do with the predicate.  Sometimes the context forces one to use, for example, the two-placed predicate (for example, in answer to the question ‘Joe is eating what?’; sometimes which predicate one invokes is purely a matter of choice.

If all of the predicates demand a certain argument (for example, ‘Joe’ in ‘Joe eats’), what is so demanded is an argument that is not also an adjunct.  If not all of the predicates demand a given argument (‘fried egg’, ‘in the kitchen’), that argument is an adjunct.  In this way, the demands criterion is rescued.

I picture the relations formed by these predicates as follows:

One-place relation formed by _eats:

EATS
PERSON_EATING
PERSON( NAME(‘Joe’) )
PERSON( NAME(‘Juan’) )
PERSON( NAME(‘Kha’) )
PERSON( NAME(‘Cliff’) )

Here the key is, of course, PERSON_EATING.  The ellipses ‘…’ indicate all the further tuples needed to make this relation satisfy the Closed World Assumption.  (The Closed World Assumption states that a relation contains all and only those tuples expressing the true propositions generated by completing the predicate with the relevant argument(s).)

Two-place relation formed by _eats_:

EATS
PERSON_EATING FOOD_ITEM_BEING_EATEN
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This fried egg’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) )

Here the relation formed by _eats_ is a subtype of the supertype formed by _eats.  That is to say, PERSON_EATING is a unique key in this relation, but it is also a foreign key to the PERSON_EATING attribute of the relation formed by _eats.

This means of, course, that in each tuple there is just one thing that the person is eating.  This constraint would be natural enough if one restricts the now of the present tense eats enough so that only one thing could possibly be getting eaten, for example, the egg one piece of which Joe is now bringing to his mouth via a spoon.  But, of course, if one stretches out this now enough so that our hypothetical author could write:   “Joe goes into the kitchen.  Joe is ravenous.  Joe eats a fried egg, an apple, and a salad,” one could not treat the one-place relation as a subtype of the two-place relation.  I think the solution in this case would be to treat what gets eaten as a meal, a meal comprising one or more items.  The meal then could be treated relationally the way an order and its order-items get treated, the orders going into one relation, and orders and order-items going into another, with the orders and order-items together comprising a unique key.

The predicate _eats_ _ (as in ‘Joe eats the fried egg in the kitchen’) can be treated the same way.  And so on for any number of possible adjuncts that a predicate might accept.

If I can get away with this move, then, an adjunct would be any argument that is 1) accepted by a predicate in which the corresponding relation is a subtype of another relation, and 2) the parameter which takes that argument corresponds to an attribute in the subtype relation which is not a foreign key of the supertype relation.  An adjunct then is one kind of argument.  Non-adjunct arguments (arguments that are just arguments, arguments simpliciter) correspond to a unique key in a supertype relation; adjuncts in turn are arguments not corresponding to any attributes in the subtype relations that are foreign keys to that unique key in the supertype relation.

Notice how this treatment of arguments vs. adjuncts (that is to say, arguments that are just arguments and arguments that are also adjuncts) corresponds to the way “optional (nullable) columns” in SQL tables get turned into actual relations, which cannot contain “null values”:

SQL Table (what is eaten is an optional or “nullable value”):

EATS
PERSON_EATING FOOD_ITEM_BEING_EATEN
Joe  Fried egg
Juan
Kha Bowl of Pho
Cliff
…

Here PERSON_EATING is a not-null column, and FOOD_ITEM_BEING_EATEN is a “nullable” column.

This looks like a single relation with an optional parameter (FOOD_ITEM_BEING_EATEN).  So if one both accepts the demands criterion and takes the  SQL table as their cue, PERSON_EATING would be an argument because it is not optional, i.e., always demanded and FOOD_ITEM_BEING_EATEN would be an adjunct because it is optional.  But then one has no way of accounting for when FOOD_ITEM_BEING_EATEN isn’t optional — for example in answering the question ‘what is Joe eating’?  (Compare with the COMMISSION column in the EMP table of Oracle’s sample SCOTT schema when the employee is a salesman.)  One would either have to try to explain away — an impossible task? — the times when eats surely seems to demand not one, but two arguments, or they would have to give up the demands criterion as the way to distinguish between arguments and adjuncts.

But of course SQL is confused.  The SQL table above is mushing together two different relations, the relation formed by _eats and the relation formed by _eats_.  Disentangle the two relations, and you get a two-fer.  You get rid of the nulls, and you also rescue the demands criterion for distinguishing between arguments simpliciter and arguments that are adjuncts.

When you disentangle the relations, you can see that what is optional, when one is talking about adjuncts, is not the attribute value (e.g., fried egg), but which predicate one invokes when they say eats.  To put it a different way, the attribute value is optional only because the predicate is.

I submit, then, that treating a verb as invoking different predicates whose corresponding relations are involved in subtype/supertype relationships does away with the confusing situation that challenges the demands criterion:  i.e., the initially confusing fact that sometimes an argument seems to be demanded for the verb, and sometimes it seems not to be.

Today’s homage to Plato’s SYMPOSIUM is Channing Tatum (aka Magic Mike) again, as in the previous post.

How can anyone get anything done with such beauty walking the earth?

The Wikipedia article Argument (linguistics) starts its discussion of the argument/adjunct distinction by asserting that an argument is what is demanded by a predicate to complete its meaning, while an adjunct is not so demanded.  For example, if someone asks me “What is Joe eating?” my answer would be drastically incomplete if I replied “eats.”  My answer would still be drastically incomplete if I supplied just one argument, ‘Joe’, to say ‘Joe eats.’  Only when I supply a second argument, say, ‘a fried egg’, would my reply not create a sense of a question ludicrously left hanging and an answer simply not given.  The predicate _eats_ demands two arguments, such as  ‘Joe’ and ‘a fried egg’ for my reply to make any sense.

( This example, of course, is my own; I am offering it (maybe tendentiously?) in order to make drawing certain conclusions more natural. )

‘[I]n the kitchen’, however, is an adjunct, since nothing would be left ludicrously left hanging in the air were I to leave that argument out of the proposition “Joe eats a fried egg in the kitchen.”  The predicate eats does not demand that argument.

This criterion — i.e., what is demanded by a predicate to complete its meaning … henceforth I will call this the ‘demands criterion’ — runs into trouble when one notices that sometimes eats demands two predicates, but sometimes demands just one.  One might say:  “Joe goes into the kitchen.  Joe eats.”  ( Imagine a novelist or short-story writer working in a certain style.)  Although one could just as well say “Joe goes into the kitchen.  Joe eats a fried egg”, the argument ‘a fried egg’ is not demanded in this particular piece of discourse.

So if one wants to maintain that the predicate eats takes two arguments, they would  have to abandon the ‘what is demanded by a predicate to complete its meaning’ criterion and find another criterion for what is to count as an argument and what is to count as an adjunct.  This a contributor (doubtlessly not the same person who put forward the ‘demands’ criterion) to the Wikipedia article cited above tries to do.

But if one wants to retain the demands criterion, they can assert that two different predicates may get invoked, depending upon the context, depending upon the circumstances, when someone utters the word ‘eats’ in a stretch of discourse.  ( I am not clearly distinguishing between predicate and word here; perhaps I don’t necessarily need to just right here.)  When one invokes the predicate in order to answer the question ‘What is Joe eating?’, invoking the predicate creates a proposition, or tuple, in a 2-place relation.  In circumstances in which nothing is left ludicrously hanging in the air when one says ‘Joe eats’, the predicate creates a proposition, or tuple, in a 1-place relation.  There are two different predicates that may get invoked when one utters ‘eats’.  And depending upon which predicate gets invoked, ‘a fried egg’ is either an argument or an adjunct.

Two-place relation (demands what is eaten to complete the meaning):

EATS
PERSON_EATING FOOD_ITEM_BEING_EATEN
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This fried egg’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This strip of bacon’) )

Here the key is composite, comprising both PERSON_EATING and FOOD_ITEM_BEING_EATEN, since we would may want to answer the question “What is Cliff eating?’ with “Cliff eats a fried egg and Cliff eats a strip of bacon.”

One-place relation (does not demand what is eaten to complete the meaning):

EATS
PERSON_EATING
PERSON( NAME(‘Joe’) )
PERSON( NAME(‘Juan’) )
PERSON( NAME(‘Kha’) )
PERSON( NAME(‘Cliff’) )

Here the key is, of course, PERSON_EATING.

Sometimes what Joe eats is a ‘core element of the situation’, sometimes it is not.  In a possible world there exists a tribe for whom the amount of  energy pounded into the ground by John’s running is a core element of the situation runs, such that something is left ludicrously hanging in the air when one simply says ‘John runs’ and not (to invent a new syntactic marker, ‘tha’, which expresses ‘the energy absorbed by the ground when John runs”’, just as ‘to’ expresses ‘the place to which John ran’ ) ‘John runs tha 1,000 <<some unit of energy>>’.

When what is eaten is an adjunct, not an argument, one can, I think, treat the attribute PERSON_EATING in the two-place relation as a foreign key dependent upon the  PERSON_EATING attribute in the one-place relation.   would be both a unique key in that relation and a foreign key to the one-place relation.  This kind of design is, of course, how one would avoids “nulls” or “optional values” in a SQL table like the following:

SQL Table (what is eaten is an optional or “nullable value”):

EATS
PERSON_EATING FOOD_ITEM_BEING_EATEN
Joe  Fried egg
Juan
Kha Bowl of Pho
Cliff
Cliff

Yes — there is a certain oddness, a certain ugliness, to having Cliff suffer from two “null values”.  Maybe there is something fishy about the SQL idea of a “null value”?  But the SQL table does convey the idea that an adjunct is an optional value, while an argument is required.  After conveying this idea, we can get rid of the SQL table with its dubious nulls and replace it with the two-place relation EATS whose PERSON_EATING attribute is a foreign key to the one-place relation.

EATS
PERSON_EATING FOOD_ITEM_BEING_EATEN IN ORDER TO
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This fried egg’) ) REASON( NAME(‘Gain Nutrition’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) ) REASON( NAME(‘Gain Nutrition’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) ) REASON( NAME(‘Gain Nutrition’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) ) REASON( NAME(‘Gain Nutrition’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) ) REASON( NAME(‘Show how macho he is’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) ) REASON( NAME(‘Show how much pain and suffering he can endure’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This strip of bacon’) ) REASON( NAME(‘Indulge in a guilty pleasure’) )

Here of course, the key is PERSON_EATING, FOOD_ITEM_BEING_EATEN, and IN_ORDER_TO.

This is the way of treating the argument/adjunct distinction that I prefer at the moment, possibly with no good argument for preferring this way to the alternative. The alternative that is at the back of my mind as I write this is something like the following:  there is only one predicate eats, which is a two-place relation.  Or rather, there is only one primary, non-derived predicate eats.  In those cases in which the what-is-eaten argument is optional (so we are giving up the demands criterion for what is to count as an argument), we are projecting on the relation EATS on the PERSON_EATING attribute, to generate propositions such as “Joe eats something.”

EATS(1)
PERSON_EATING SOME ATTRIBUTE
PERSON( NAME(‘Joe’) ) Some thing or things
PERSON( NAME(‘Khadija’) ) Some thing or things
PERSON( NAME(‘Juan’) ) Some thing or things
PERSON( NAME(‘Kha’) ) Some thing or things
PERSON( NAME(‘Cliff’) ) Some thing or things

Here I envisage the demi-urge performing the needed projection by ignoring the FOOD_ITEM_EATEN attribute (perhaps even forgetting there is such an attribute in the relation), then, in order to avoid duplicates (we don’t want our demi-urge to be seeing double!), collapsing what had been two appearances of Cliff into just a single appearance.

The picture of relations above may be pretty (forget the picture of the SQL table — that is definitely not pretty…nothing connected to SQL ever is), but even prettier is  Channing Tatum aka Magic Mike, who is today’s homage to Plato’s SYMPOSIUM:

Notwithstanding all of my rapturous sighs at the moment, my sole interest in Magic Mike is, of course, as a stepping stone first, to the Relational Algebra, and then, ultimately, to the Platonic Form of Beauty.

Logical Pairings

In previous posts I’ve tried to interpret the canonical Tagalog sentence (e.g., maganda si Taylor Lautner) in terms of an equality relation, GORGEOUS_EQUALS_GORGEOUS.  Conceptually, the relation is formed by logically pairing each member of the set GORGEOUS (MAGANDA) to each of the members, then taking a subset of the set that results from this logical paring.  That subset comprises those logical pairings in which each member of the pair is identical with the other.

What do I mean by ‘logical pairing’?  In the real world, to pair one thing with another is to bring the two things together in some way.  One may pair, for example, some particular matte board, with its particular color, with the painting one is getting framed.  Here, the matte board and painting are getting physically paired.  Or one may pair John with Bill by picturing them in the mind’s eye as together as a couple.  Or one may pair John with John by first seeing him double (i.e., seeing him twice but simultaneously), then by realizing the two Johns are in fact one.

To get a logical pairing, abstract from any concrete form of pairing, that is, ignore any particular way in which the bringing together is done.  Ignore in fact everything about them except that they go under the heading ‘bringing together’ (since maybe that is the only single thing they all have in common.)   Then be content with the fact that, while each member of the set MAGANDA can potentially be brought together with every member of that set,  any actual pairings will be performed just every now and then, and only for a few members.  (For example, in a particular article, Dan Savage pictures Ashton Kutcher and Matt Damon together.)  A logical pairing is a bringing together in which all concrete details of the bringing together (how it is done, in what sense the things are brought together?  Physically?  In the imagination only?  By already knowing that the “objects” of one’s double vision are in fact one and the same?) are ignored.  One salient detail in particular is ignored:  is the pairing actually being done in any given instance, or is it just something that could be done?

If one does not want to rest content with each member of the set being brought together just potentially with every other member of the set, they (plural third person intentionally being used here as a neutral singular third person) are free to imagine a Demiurge ala Plato or a God ala the medievals whose cognitive capacities are sufficiently large as to simultaneously bring together in its mind’s eye every member of the set MAGANDA with every member of that set, so large, in fact, as to be able to see Matt Damon twice with the mind’s eye but already know that Matt Damon is, well, Matt Damon.

I will end by confessing that I like to think of projection as the Demiurge’s ignoring one or more attributes of a relation, and of restriction as the Demiurge’s ignoring one or more tuples in the relation.

Today, my homage to Plato’s SYMPOSIUM (first, gorgeous guys, then the Relational Algebra, then the form Beauty itself) will take the form of a concrete (not just a logical) pairing of Matt Damon and Ashton Kutcher:

Sigh.  There is too much beauty in the world.

Some More Clean-Up Work: Propositions And States Of Affairs

Following Chisholm, I have been identifying propositions with states of affairs.  A proposition is a subset of the set of states of affairs.  The state of affairs of John grasping a doorknob at time t_0 in Chicago is a state of affairs that always occurs (or always fails to occur).   States of affairs like this one are propositions.  The truth (falsity) of a proposition is nothing but a certain state of affairs occurring (failing to occur).  I am ignoring the question, which is pestering me right now, of why then it seems so awkward to talk about a ‘true’ (‘false’) state of affairs.  From The Stanford Encyclopedia of Philosophy article on Roderick Chisholm:

Consider the state of affairs that is expressed by the sentence ‘Someone is walking’. Chisholm wanted to say that this state of affairs occurs whenever someone walks, and fails to occur at times when no one is walking. Other states of affairs are not like this. For them, it is impossible to sometimes occur and sometimes fail to occur. Chisholm claims that this provides the opportunity for an ontological reduction. We can define a proposition as a state of affairs of this latter sort — it is impossible for there to be times when it occurs and other times when it does not occur. A true proposition is thus one that occurs; and afalse proposition is one that does not occur. Chisholm thinks that we may understand the principles of logic to be about these propositions. By saying that a fact is a true proposition, Chisholm gains yet another ontological reduction ([P&O], 123).

Chisholm thought that in some cases it makes sense to speak of the location at which a state of affairs occurs. Suppose John walks in Chicago at a certain time. Then Chisholm would be willing to say that the state of affairs of John’s walking occurs in Chicago and at that time.

Those states of affairs that are not propositions are events.   I am going through this stuff a bit impressionistically; the chances of my making an error someplace are high.

The tuples in the body of a database relation are propositions.  That is to say, they are states of affairs.  In a conventional database, these are always states of affairs occurring now, and now, and now…. John is an employee of WIDGETS_R_US now, the ‘now’ being implicit in the presence of that tuple in the relation.   In a temporal database as described by Date and Darwen (TEMPORAL DATA AND THE RELATIONAL MODEL),  these are states of affairs that occurred during a time period, or are occurring now (“Since t_0….”), the relevant time periods being explicitly stated in the tuple.

Since propositions are nothing but states of affairs of a certain kind, the operations of the Relational Algebra are operations on states of affairs of that kind.  On the relation ‘Standing_To_The_RIGHT_Of’, for example, we can perform a RESTRICT operation that delivers to us the state of affairs of Don standing to the right of Genghis Khan, then perform a PROJECT operation on that derived relation to obtain just Don.

We will figure out later what to do with Don now that we have him.

My homage to Plato’s SYMPOSIUM for this post will be Matt Damon.  This time we are a bit further along on the way towards eros for mathematical beauty:

But let’s not forget it all originally stems from eros for gorgeous young men.

Semantic vs. Syntactic Arguments

In a previous post, playing the role of a would-be ordinary-language philosopher working in Tagalog (which language, to the the total consternation and perplexity of the Spanish grammarians during the 1600s, lacks the verb ‘to be’), I tried to provoke the suspicion that there is no single relation IS that could be pictured as follows:

IS (0)
THING PROPERTY
NUMBER( NAME(‘3’) ) PROPERTY( NAME(‘Prime’) )
CAR( NAME(‘Car With Serial Number 1235813’) ) PROPERTY( NAME(‘Red’) )
FLOWER( NAME (‘Rose With Barcode 3185321’) ) PROPERTY( NAME(‘Beautiful’) )
MINERAL_ITEM( NAME(‘Grain Of Salt Mentioned By Hegel’) ) PROPERTY( NAME(‘Cubical’) )
MINERAL_ITEM( NAME(‘Grain Of Salt Mentioned By Hegel’) ) PROPERTY( NAME(‘White’) )

Contra John Duns Scotus, for example, there is no single relation ‘contracts’ holding between a universal existing as always-already contracted into a particular. (To back up for a moment, the property denoted by NAME(‘three-sided’)is the universal denoted by NAME(‘three-sidedness’)existing as already contracted into a particular triangle.) Nor is there any other single relation which we can identify with the verb ‘is’. Or … letting my Tagalog ordinary-language suspicions run wild for the moment … so I will suppose.

There is no semantic relation (we are supposing) between the particular thing and the particular property.  But there is a syntactic relation between two names, pictured as follows;

IS (0)
THING PROPERTY
NAME(‘3’) NAME(‘Prime’)
NAME(‘Car With Serial Number 1235813’) NAME(‘Red’)
NAME(‘Rose With Barcode 3185321’) NAME(‘Beautiful’)
NAME(‘Grain Of Salt Mentioned By Hegel’) NAME(‘Cubical’)
NAME(‘Grain Of Salt Mentioned By Hegel’) NAME(‘White’)

Voila:  here is the distinction between semantic vs. syntactical arguments to a verb aka predicate that puzzled me in an earlier post.  NAME(‘3’) and NAME(‘Prime’) are syntactic predicates to the verb/predicate ‘is’.  NUMBER( NAME(‘3’) ) and PROPERTY( NAME(‘Prime’) ) are the semantic predicates to the verb ‘is’ — or would be if there were such a verb ‘is’ that took semantic arguments.

In the spirit of ‘let’s see how long I can get away with this’, let me propose the following chain of events for verbs such as eats that do take semantic arguments.  Consider a relation like the one pictured here:

EATS (0)
NAME_OF_PERSON_EATING NAME_OF_FOOD_ITEM_BEING_EATEN
NAME(‘Joe’) NAME(‘This egg’)
NAME(‘Juan’) NAME(‘This fajita’)
NAME(‘Kha’) NAME(‘This bowl of Pho’)
NAME(‘Cliff’) NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’)

When used in ordinary discourse, rather than mentioned as sentences with whatever syntactic properties, these tuples with their syntactic arguments get transformed into the following tuples with their semantic arguments:

EATS (0)
PERSON_EATING FOOD_ITEM_BEING_EATEN
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This egg’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) )

If we think of the intransitive and transitive verbs ‘eats’ as really being the same verb on the semantic level (though it is not clear to me that they are the same), ‘Joe eats’ would be ‘Joe eats something’.  We can derive the corresponding tuple from  the EATS relation first by projecting on the attribute PERSON_EATING:

EATS(1)
PERSON_EATING FOOD_ITEM_EATEN
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This egg’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) )

Then by doing a RESTRICT on Joe:

EATS(2)
PERSON_EATING FOOD_ITEM_EATEN
PERSON( NAME(‘Joe’) ) FOOD_ITEM( NAME(‘This egg’) )
PERSON( NAME(‘Khadija’) ) FOOD_ITEM( NAME(‘This souffle’) )
PERSON( NAME(‘Juan’) ) FOOD_ITEM( NAME(‘This fajita’) )
PERSON( NAME(‘Kha’) ) FOOD_ITEM( NAME(‘This bowl of Pho’) )
PERSON( NAME(‘Cliff’) ) FOOD_ITEM( NAME(‘This plate of Thai food with a 5-star Thai-spicy rating’) )

The existence of verbs that are sometimes transitive, sometimes intransitive  is what motivated Santorini’s distinction between semantic and syntactic arguments to a verb.

Although I have labored over making this distinction for an embarrassing amount of time, it becomes quite easy to make once one has the notion of a SELECTOR available as a conceptual tool.

I’d like to mention as a final note that possibly we should think of the arguments of eats as always already nested inside the selectors PERSON and FOOD_ITEM; in other words, the names are always transparent, so to speak, letting us see through them the persons and food items, the semantic arguments, named.  It is only under special circumstances — say when the transparent denoting function of the sentence breaks down … maybe one has been staring at the sentence for too long — that the selectors PERSON and FOOD_ITEM get stripped away and we see the names, the syntactic arguments, doing the denoting.  (The sentences always seem to be breaking down this way for C.J. Date in his article “SOME OPERATORS ARE MORE EQUAL THAN OTHERS” in his LOGIC AND DATABASES:  THE ROOTS OF RELATIONAL THEORY.  I get the funny feeling that for him a sentence or expression functions normally at first, but when he stares at it too long it suddenly loses its transparency and becomes an opaque relation between names.  See pages 42 and 45, and see if you get the same impression.)  This final note has been brought to you by the balefully compromised spirit of Martin Heidegger, which was nagging me as I wrote the above.

And now, in the spirit of Plato’s SYMPOSIUM, I would like to picture something a little less dry than the pictures of Relations shown above.  Today it is Kellan Lutz who is serving as my stepping stone, first, to the Relational Algebra, then, finally, to the form of Beauty itself:

(Rapturous sigh.  How can one live in this world when there is so much beauty in it?)

Update 12/16/2012:  Corrected some errors in the names of some of the Relations; tried to improve the flow of the writing.

Selectors And Semantic vs. Syntactic Arguments

In case anyone wonders (“feel free to come to the point when you finally decide what it is”), the point of the following ramblings is to arrive at a place where I can make a distinction between semantic arguments and syntactic arguments.  The point of making this distinction will become clear (or not) in a later post.  Making the distinction is part of my attempting to put in my own words the argument that Tagalog lacks a subject.

In the previous post, I argued (or claimed, or made the completely unsupported, nay, spurious assertion, as the case may be) that the semantics of Maganda si Robert Pattinson can also be given by the following statement in the database language Tutorial D:

GORGEOUS_EQUALS_GORGEOUS{THIS_ONE, THAT_ONE} where THIS_ONE = PERSON(NAME(‘Robert Pattinson’))

This statement includes the Selector PERSON(NAME(‘Robert Pattinson’)).  Let me unpack a bit what this is. Before I start, I’d like to point out that I THINK that it is  legal in Tutorial D to nest one selector inside another…

NAME(‘Robert Pattinson’) is a operator or function that takes the string ‘Robert Pattinson’ and selects one and only one name.  I will take the concept ‘selects’ as primitive here.  Any implementation of this selector in a physical computer would involve shuffling around ones and zeros until the computer spits out, i.e., returns, one member of the set NAME.  NAME would include strings, but subject to certain limitations.  For example, I assume a  name would have to be, at least, less than 1 billion characters long.  NAME would also include more than strings (that is, representations of text):  a name can be selected by a sound.  So NAME(<<some representation of a sound>>) could also select the name Robert Pattinson. (The reader will notice that I have not yet decided on how to represent, in the absence of a formal selector, a name as opposed to a string as opposed to the person himself…)

PERSON(NAME(Robert Pattinson)) would take the name selected by NAME(‘Robert Pattinson’) and return a member of the set PERSONS, i.e., Robert Pattinson himself.  I don’t know how a computer would implement this operator, but a human being would be implementing that operator in the following type of circumstance:  say, I am sitting in a restaurant.  Someone in the table next to me says:

I hereby officially declare myself to belong to Team Edward because Robert Pattinson is just too gorgeous.

One part of that utterance, the part that I hear as the word ‘Robert Pattinson’, is the end point of a long causal chain that begins, say, when the parents of Robert Pattinson, after endless wrangling and indecision, finally agree to call their baby ‘Robert’; the doctor in the Maternity Ward crosses out the ‘baby boy’ in ‘baby boy Pattinson’ and writes in  ‘Robert’ on the birth certificate (call this the ‘baptismal event’) … endless events … a director or producer chooses the person named by ‘Robert Pattinson’ to play Edward Cullen in TWILIGHT … endless events…the person sitting at the table next to me sees TWILIGHT…he reads in a magazine he buys at the supermarket that Robert Pattinson played the part of Edward Cullen…he emits a set of soundwaves at the table next to me, which in turn trigger God-only-knows what processes in my brain, until I hear ‘…Robert Pattinson….’  That entire causal chain, ending up in the wetware of my brain, selects the person Robert Pattinson.  THAT’s the implementation of the selector PERSON(NAME(<<some representation of certain sound waves>>)).  Speaking metaphorically and a bit picturesquely, the selector spits out, or returns, Robert Pattinson himself, the flesh-and-blood Robert Pattinson who lives in (I would say ‘Valencia, California’, but that is where Taylor Lautner lives)…. Speaking literally, the selector selects Robert Pattinson himself.

(See Saul Kripke, who apparently never explicitly endorsed this causal theory of reference aka selection.  Gareth Evans would apparently deem this theory, as stated by me, to be naive, but it seems perfectly intuitive to me.)

Invocations of selectors produce literals (more accurately, I guess, are literals).  So whatever else Robert Pattinson himself may be, he is a literal value.

Let me take the liberty of allowing selector invocations as arguments supplied to the parameters of functions, so that we can replace x with the argument PERSON(NAME(‘Robert Pattinson’)) in the function x EQUALS x to produce a true proposition.  Below, I have identified, ala Chisholm, propositions with states of affairs in the world:  here, with Robert Pattinson being identical with Robert Pattinson.  This proposition gives us the semantics of the utterance “Robert Pattinson equals Robert Pattinson.”

I will therefore call the invocation of PERSON(NAME(‘Robert Pattinson’)) a semantic argument.  By contrast, the invocation of NAME(‘Robert Pattinson’), occuring inside an utterance, spoken or written, is a syntactic argument.  In this way, I make sense of the semantic arguments vs. syntactic arguments distinction I puzzled over in a previous post.

I do not know, of course, whether this is the distinction that Beatrice Santorini wanted to make.

I will end by making another homage to Plato’s SYMPOSIUM, according to which interest in Robert Pattinson, Taylor Lautner, Kellan Lutz et al ultimately leads to interest in the Relational Algebra, and from there, to the Form of Beauty itself:

Wow, I love that slightly-unshaven look…(the reader may  hear a rapturous sigh…)

Now, having briefly lapsed into a lower form of eros, I will go back to eros for the Relational Algebra in connection with Semantics….

Update:  After hitting the publish button, I saw this quote from the first Jewish Prime Minister of Great Britain:

The best way to become acquainted with a subject is to write a book about it.

Benjamin Disraeli

Or blog about it at length.