Earliest Reference to "Ghoti"

[Benjamin Zimmer]

On 12/12/06, Benjamin Zimmer <bgzimmer at ling.upenn.edu> wrote:
> On 12/12/06, Laurence Horn <laurence.horn at yale.edu> wrote:
> > I remember hearing a no doubt ghoti-y tale that the observation was
> > due to G. B. Shaw.  A little searching indicates that it is a
> > widespread canard (esp. good on toast).  See e.g.
> > http://alt-usage-english.org/excerpts/fxwhat04.html
> > (In this piece, Jim Scobbie doesn't attempt to date the "ghoti" fish-tale.)
>
> For further on this, see my alt.usage.english/sci.lang post from June 2004:
>
> http://groups.google.com/group/sci.lang/msg/5501afa30de85a88
>
> Executive summary: "ghoti" shows up unattributed in 1938 (1937 thanks
> to Fred's new cite), then directly attributed to Daniel Jones in 1943,
> then vaguely attributed to G.B. Shaw in 1946 by the canard-alicious
> Mario Pei (who perhaps was getting Jones and Shaw confused).

There's another 1938 cite on Google Books, from Albert Galloway
Keller's _Brass Tacks_ (p. 16): "It is more than a quip when someone
offers 'ghoti' as the way to spell 'fish': 'gh'
as in 'rough,' 'o' as in 'women,' 'ti' as in 'attention.' This is only
in snippet view, so there's no way of getting more context, but it
seems to have the same unattributed status as the other cites from
1937-38.

I see Stephen Goranson has already posted the "stage-Frenchman"
version from 1943. There's also an attribution to Daniel Jones from
1946 in _Speech_ by Andrew Thomas Weaver & Gladys Louise Borchers
("Some time ago Daniel Jones, a distinguished authority on phonetics
at the University of London, suggested a ludicrous spelling of
fish...") Earliest Shaw attribution on Google Books is unsurprisingly
from Mario Pei, in _The Story of Language_ (1952). That book was
probably the biggest single vector for the canard, more so than Pei's
1946 syndicated language columns.


--Ben Zimmer

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org