"23 skiddoo"

[Douglas G. Wilson]

> > I doubt the "poison" sense here. I tentatively prefer a laxative powder as
> > the original sense.
>
>Again, what is your evidence?  And if there is no evidence, what basis is
>there for a preference?
>
>Fred Shapiro

Well, I said "tentatively" after all. Of course there is some evidence,
although I've seen none which I know to be decisive: thus "tentative".

In the early 20th century and earlier, I believe "take a powder" had a
transparent conventional meaning like "take a pill" does today, viz., "take
a dose [presumably of medication]". There were many kinds of powders, but
the pharmacopeia was quite narrow by modern standards: laxatives,
analgesics, and soporifics were commonplace powders. So if one said "I
couldn't sleep, so I took a powder" one would understand this to refer to a
soporific; today it would be "... so I took a pill" more often, I guess. In
the early days of "take a powder" = "depart", maybe 1920 or so, there was
the variant "take a run-out powder" meaning "run out" = "depart": this is
analogous to the modern "take a chill pill" = "chill", i.e., "cool
off"/"calm down". Three types of powder have been proposed here: poison
(e.g., Cohen), magical powder which makes one disappear (e.g., Chapman),
laxative powder (e.g., Partridge). Of these, the laxative possibility best
accounts for the "run-out" connection IMHO (BTW, I tentatively take "run
out" as "run out to the toilet" rather than referring crudely to a liquid
trickling out). A completely speculative possibility: Tarrant's Aperient [a
common laxative powder ca. 1900] > "Tarrant powder" > jocular "tear-out
powder" [meaning one must tear out to the toilet after the dose] > "run-out
powder" [with double-entendre].;

In the absence of direct evidence, one can employ reason. For example, I
know of no strong evidence for the derivation of "chad", but derivation
from a Scots word for "gravel" is a reasonable tentative concept while the
acronymic origin is implausible and the "Mr. Chadless" etymology is
laughable (IMHO). That's why Fred Shapiro put forth the "gravel" hypothesis
in "Verbatim" in 2001. Another example which I've exposed here: "poontang".
Is it from French "putain" (US ca. 1910)? Seems reasonable, but where is
the evidence? I know of none, and I can think of four other plausible
etymologies ... but it's good enough for a tentative guess, for the OED,
and for M-W, and for me too.

When there is 'evidence', there is often still room for considerable doubt.
For example, Popik and Cohen have presented texts supporting the name
"Jinks" as the ancestor of the word "jinx" = "curse", but I do not agree
that the evidence is at all convincing (see recent posts on this list).
Read presents very extensive textual support for his "OK" derivation (which
I find reasonably convincing), but not everyone is satisfied. Etc.

-- Doug Wilson