[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index

Re: The notion of semantics in Rosen's writings



Yay, Tim!
 
In your comment: "One of the common themes in Rosen's thinking is to remove unjustified a priori restrictions wherever they occur," you really nailed down clearly one of the most salient aspects of his approach to science (and everything else, really). In fact, I think this was just a part of his nature from very early childhood. Perhaps this is what comes from being a child prodigy who also happens to be the only child of two very ordinary immigrant parents-- who both had large immigrant families that each came to the US as a unit-- and who were both trying to assimilate. They really didn't know what to make of him. I'm sure they tried to cope with his active mind by trying to curb him and no doubt they issued warnings and edicts based on them... which he (no doubt) tested and found the warnings were groundless. If "P, then Q" works in the negative, too! "If not P, then not Q" can be turned into a new encoding: "If the warnings are groundless, then the parental edicts are baseless."
 
He said he figured out very young that it was better not to ask permission to do something he wanted to do; it was better to just do it. If you ask for permission, and it is withheld based on inapplicable fears (which they would never concede)... then you have two choices: you can either accept their decision, which doesn't seem fair... or you can do what you wanted to do anyway. But once you have asked permission, if you go against a parental edict, you have committed a much more serious sin! You are insubordinate! Off with your head! If you just do it, without asking permission, they may get mad after the fact, but then you just apologize and that tends to settle everyone down.
 
In using this approach with science, though, he didn't apologize. Probably because he considered everybody in that milieu as equals and therefore he owed no fealty or familial respect to people who were trying to rein him in based on their own fears. I would say that the only person he paid a great deal of familial respect to was Rashevsky (his PhD advisor/mentor). In many ways, Rashevsky was the father he never really had.
 
Judith
PS: Incidentally, I confess that  I have often been irked by the numerous times, in my father's published work, where he gave far more credit to Rashevsky than was actually warranted. I questioned him about this, once, and he said that since Rashevsky had arrived at a similar conclusion (that whatever scientific law was not applicable), prior to Dad being in Chicago, so he felt that Rashevsky deserved the credit for the insight. My argument was; "Yeah, but Dad... YOU not only arrived at the insight on your own, but you developed it. Why are you giving HIM all the credit for it?" He just shrugged and couldn't really give me an answer that made sense to me beyond that it was his own sense of duty and ethics (and sentimentality, in my view). Dad actually loved "the old man" as he called Rashevsky. And Rashevsky was shafted, big time, when Lewontin came to Chicago, etc. I think my father always felt he had kind of left Rashevsky unprotected, by moving on to Buffalo to pursue his own career.
----- Original Message -----
From: Tim Gwinn
To: ***
Sent: Sunday, August 21, 2005 8:54 AM
Subject: Re: [ROSEN] The notion of semantics in Rosen's writings

Carlos,

I know Judith has already replied, but sometimes multiple replies are
helpful. :)

> The first problem I have is with the definition of
> 'encoding' and 'decoding'. Rosen defines in the book
> Life Itself encoding and decoding as the 'translation'
> between 'formalisms'. I understand this as the idea of
> translating well-formed formulas given in a formal
> language to well-formed formulas of another formal
> language. But, what is 'encoding' and 'decoding'
> between a formal language and a 'natural system'?


TG: Encoding is the mapping of a phenomenon (i.e., an observable) in the
natural system to a counterpart in the formal system.This is associated with
the notion of measurement, where measuring some physical observable results
(typically) in a number, a formal thing.
Decoding is the inverse operation, mapping a formal entity to an observable
in the natural system. This is associated with the notion of prediction,
whereby we use results from a formal model to predict what some value of
some observable would be if we were to measure it.

So, for example, if we measure the current in a physical electrical circuit,
we are encoding that observable into a number, such as "10 amps". If we then
use our model of electrical circuits, we can say that "if we double the
resistance in the circuit, the current should be cut in half to 5 amps,
according to our model". This is decoding the formal results back to the
physical system - making a prediction about it based upon our model.

There are much more extensive descriptions of the modelling relation in the
earlier books "Anticipatory Systems" and "Fundamentals of Measurement".



> Also, is not very clear to me what qualifies as a
> 'natural system', since Rosen defines them as systems
> "in the ambience or external world".


TG:  A natural system is simply a collection of observables from the world
outside of us. A 'car' or a 'cat' or a 'proton' is a natural system,
comprised of the observables we have chosen to consider as belonging to that
system. Again, the books AS and FM go into much more detail.


>  I can't
> understand Rosen's point very well, so I'd pleased to
> know in what exact sense" the entailment structure of
> the language itself can change by virtue of what the
> language is about" and if Rosen effectively is
> asserting this.


TG: His statement is: "...we leave open the possibility that the entailment
structure of the language itself can change by virtue of what the language
is about-that is, that it can depend on semantic as well as syntactic
features." That is, there is no a priori reason to exclude such
possibilities. One of the common themes in Rosen's thinking is to remove
unjustified a priori restrictions wherever they occur.


>  Are the logical constants of the
> formal language of a formal logic a candidate for
> "entailment structure"?


TG: 'Entailment structure' refers to relationships between entities, not to
entities themselves. Can you give an example of what you mean?


Regards,
Tim