Watkin's IE dictionary

[Jim Rader]

----------------------------Original message----------------------------
I'm sending this to the list rather than to Larry alone since it may
be a matter of wider interest.

When the first edition of the _American Heritage Dictionary_ was
published in 1969, the list of PIE roots was printed as an addendum
at the back of the book.  The list was maintained as an addendum in a
desk dictionary  (the "New College Edition") published in 1976 (at
least I believe it was--I don't have a copy at hand).  When _The
American Heritage Dictionary, Second College Edition_ was published
in 1982, however, the PIE roots were dropped as an addendum and
published as a separate book, _The American Heritage Dictionary of
Indo-European Roots_, which came out in 1985.  As far as I know, this
is the only occasion when this part of the dictionary was ever published
in separate form.  It's out of print, though I imagine copies of it
could be found through WWW search services such as Amazon.com's used
book department.  The scuttlebutt in reference publishing was that
the exclusion of the list was not popular among customers.  When
_The American Heritage Dictionary of the English Language, Third
Edition_ was published in 1992, the PIE roots were restored to the
back of the book as an addendum, and they have been maintained in
_The American Heritage College Dictionary, Third Edition_, published
in 1993.  (The assignment of edition numbers is confusing--there
never was an "American Heritage Dictionary of the English Language,
Second Edition," because, for reasons known only to the publisher,
Houghton Mifflin,  "second edition" was assigned to the re-edited
desk dictionary published in 1982.)

To those not acquainted with American dictionary publishing, a word
about terminology might be appropriate.  A "desk dictionary" is more
or less the size of a conventional  book that can be easily pulled
off the shelf with one hand.  It retails in the U.S. for $21-$25 or
thereabout.  All four of the major desk dictionaries put out by U.S.
publishers have the word "College" or "Collegiate" in their title
(the latter is a trademark of Merriam-Webster, which published the
first "Collegiate" dictionary in 1898, and more or less launched the
desk dictionary business). _The American Heritage Dictionary of the
English Language_ is larger and bulkier and sells for a good deal
more: $50 retail.  Its only direct competitor in size and price is
the recently published _Encarta World English Dictionary_, published
in the U.S. by St. Martin's Press and bankrolled by Microsoft.  Above
these two in size and price are the so-called "unabridged"
dictionaries, of which only two are worth mentioning, _Webster's
Third New International Dictionary_, published by Merriam-Webster in
1961, and _Random House Webster's Unabridged Dictionary_, originally
published as _The Random House Dictionary of the English Language,
Second Edition Unabridged_ in 1987.  Both of these dictionaries are
too large to be pulled off a shelf with one hand and are usually
placed on dictionary stands in American public and academic
libraries.

When the list of Indo-European roots was prepared for the first
American Heritage dictionary, Calvert Watkins enlisted the services
of  a number of people at Harvard in the '60's, mostly graduate
students as I've heard the story.   Among this group some names stand
out for work later done on their own:  I mention Bruce Bolling, Ives
Goddard, Jay Jasonoff, Peter Jorgenson, Michael Silverstein, and
Robert Underhill as names I immediately recognize, though others may
be equally well-known.  I have no idea how much input Watkins himself
had on the roots list--only he could answer that question.  It's
obvious, though, that it leans very heavily on Pokorny if you spend
any time working with it.  There are patent drawbacks to compiling a
list of Indo-European roots that can only use as comparanda words
that happen to exist--either by inheritance or borrowing--in English.
 You can't draw in comparanda from languages virtually without
reflection in the English lexicon, such as Lithuanian or Hittite.
The American Heritage style also has no means of showing roots
attested, say, only in Germanic, or only in Ingvaeonic, or only in
Old English.  The 1st edition does say for English "follow," for
example, "...Old English <folgian> and <fylgan>, from Germanic
<fulg-> (unattested)"; the 3rd edition, however, simply ends at "Old
English <folgian>."  The 3rd edition of the American Heritage
Dictionary as a whole has considerably reduced the number of
Indo-European bases cited in the Addendum compared to the 1st
edition.  Some of the trimming, most likely done for reasons of
space,is justifiable, e.g., the dropping of "<ger->, curving,
crooked; hypothetical base for a variety of Germanic words with
initial <kr->," where Pokorny has by the unconstrained use of root
extensions lumped together a great jumble of heterogeneous Germanic
etyma.  Other omissions are surprising, though; "apple," for example,
is not carried past Old English in the 3rd edition, whereas the first
reconstructed <abel->.  I have no idea how much input Prof. Watkins
had on the editing down of the root list--again, only he could answer
that question.

A very long-winded answer to Larry's question, but I hope that these
observations from someone inside the American dictionary business may
be of passing interest to list members.

Jim Rader


> The first edition of the American Heritage Dictionary carried a
> dictionary of the PIE roots attested in English, compiled by Calvert
> Watkins.  Can anybody tell me if that dictionary can be bought
> separately, and, if so, where?
>
>
> Larry Trask
> COGS
> University of Sussex
> Brighton BN1 9QH
> UK
>
> larryt at cogs.susx.ac.uk
>