🚀 go-pugleaf

In article <1992Nov1.082333.44291@kuhub.cc.ukans.edu> miner@kuhub.cc.ukans.edu writes:
>In article <1992Oct26.215238.102458@Cookie.secapl.com>, frank@Cookie.secapl.com (Frank Adams) writes:
>
>[most of earlier discussion with up to four embeddings deleted for
>convenience]
>
>Miner:
>>>>>The field I know best is Linguistics.  In this field, we make the
>>>>>following assumption:
>>>>>
>>>>>	(a) All of the languages ever spoken have had the same properties
>>>>>            as languages we have examined.
>
>[...]
>
>Miner:
>>>My examples of
>>>scientific assumptions have to be in statement form, because I have
>>>to talk about truth.  That is, I need to be able to be clear about
>>>what the propositional content of the assumption is.
>
>[...]
>
>Adams:
>> I wonder if your (a) is really most accurately presented in statement form.
>> There are, after all, some good reasons for thinking it *isn't* actually
>> true; and, as you note, even those who state it may not really believe.
>> I would like to seriously suggest that an imperative formulation of (a)
>> might more accurately represent it.  For example,
>> 
>>   (a') Don't assume that languages we can study from the past have different
>>        properties from those of languages we have examined.
>> 
>> It seems to me that a speaker who does *not* actually believe in (a) but
>> says it anyhow must really have an interpretation like (a') in mind.
>[...]
>Secondly, isn't it really deontic modality you are gravitating toward
>here?  In other words, in the end aren't you really going to end up
>saying that scientific assumptions are of the form "One ought (not) to do
>X"?  If so, it might be best to short-circuit the excursion through
>directives and go right to the question of seeing these things as part
>of the hoary old problem of getting an "ought" from an "is" (or at
>least dealing with "oughts").

Yes, I think you're right.

At that point, the derivation of (a) doesn't seem to me to be very
mysterious.  We aren't really going from an "is" to an "ought".  There is a
pre-existing "ought":

   (c) One ought to do what one can to further the science of Linguistics.

(c) is rather obviously aimed at linguists, as (a) is.  And the derivation
of (a) from (c) has already been stated: not assuming (a) produces useless
theories.

And (a) is still in some sense "falsifiable": the arguments that led us to
(a) from (c) depend on other things about our state of knowledge.  If we can
come up with an information source/methodology for dealing with changes in
the fundamental structure of language, (a) will no longer be good advice.