Re: Definition of heat

[John Denker <jsd@MONMOUTH.COM>]

At 12:42 PM 9/12/99 -0500, Carl C. Gaither wrote:
> Is Carnot's concept of heat no longer valid?

That's a well-aimed question.  It is exceedingly sad that the question even
comes up, but it does.

[And thanks for the Carnot quote.]

At 08:03 PM 11/25/99 -0700, Jim Green wrote:
> I don't see here any agreement of just what deltaQ
> might be except for the cylinder and hot plate case.

I agree that there is a deplorable lack of agreement as to the definition
of heat.

 *) The first definition (heat = thermal energy) is/was used by Carnot
(1824), Boltzmann (1872), Feynman (1963), and every experimental physicist
I've asked about it (including people with PhDs from California, Cornell,
Duke, MIT, and Stanford).

 *) The second definition (heat = heatflow) was can be traced as far back
as Slater (1939;  Dover reprint 1970) and was advocated by Zemansky in his
1970 _Physics Teacher_ article.  This definition has made it into a number
of textbooks and even into the current version of the Encyclopedia Britannica:
  http://www.britannica.com/bcom/eb/article/6/0,5716,40566+1,00.html

 *) One can dig up additional even-more-unorthodox definitions, but they
are not worth discussing.

====================================

Short-term recommendation:  The Internet Engineering Task Force has a
design rule: "Be conservative in what you produce, and liberal in what you
accept."

Specifically, in this case:

  A) When reading a document, it is usually easy to recognize which of the
two main definitions is being used.  If the first definition is being used
and you don't like that, don't get mad, just replace the term "heat" with
the term "thermal energy" everywhere.  Conversely, if the second definition
is being used and you don't like that, just replace the term "heat" by
"heatflow" everywhere.

  B) When writing a document, it would be smart to spell out what
definition you are using, and to acknowledge that other definitions can be
found in the literature.

=============

Longer-term recommendation:  The heatflow definition has got to go.

In my judgement
 -- The technical arguments in its favor are invalid.
 -- It has no practical advantage when doing thermodynamic calculations.
 -- It has no pedagogical advantage, and has serious pedagogical
disadvantages.