The Science Book, edited by Peter Tallack (2001, London:
Weidenfeld and Nicolson), a
book about the major discoveries in science, recognizes 1957 as the year
that Chomsky set out to "show that language is a skill that human beings
are innately predisposed to acquire". This is one of the 250 episodes
from the history of science that The Science Book presents.
[Angelika Kratzer]
This issue of WHISC celebrates collaboration, in particular, the
numerous
and diverse collaborations that UMass linguists were and are involved
in.
The celebration takes the form of a series of collaboration
graphs.
What counts as a collaboration? This is a
tricky
question.
For this issue, we say that
the collaboration relation holds between two individuals a
and b just in case
a cowrote a published paper with b; or
a presented a paper with b; or
a coedited a published volume with b; or
a and b were coinvestigators on a grant
together.
The definition of published must be broad enough to include
Internet
publication. It would be a mistake, here on the verge of 2004, to
restrict attention to print media, especially given
the uneasy relationship
between scientists and the world of academic publishing.
The graphs are probably not complete.
Suggestions for augmenting them are most welcome.
Send them to
.
In the
UCSC Department of Linguistics,
they are rightly proud of the fact that their faculty collaboration
graph
is connected: from any tenured professor in the department, there is a
path to any other tenured professor in the department that consists
entirely of UCSC-Linguistics-tenured-professor links.
Check it
out.
We at WHISC were curious about the collaboration graph of the UMass
Linguistics tenured
or tenure-track faculty. It turns out not to be connected (as least as
far as we know). Here is the graph we compiled; clicking on an edge
will take you to a webpage or PDF file that documents the
collaboration in question:
Kratzer
Partee
Bach
Roeper
Frazier
Kingston
Selkirk
McCarthy
Johnson
Pater
Potts
Speas
Woolford
UPDATE (Jan 1, 2004)
As you can see, the graph turns out to be more connected than
originally reported.
The Psycholinguistics Training Grant
established the Kingston--Roeper and Selkirk--Roeper links.
Why is the UMass faculty-collaboration relation
relatively small? The UMass and UCSC Linguistics departments are
about the same size, and both departments consist entirely of very
active researchers. The sparseness of the above graph is initially
puzzling.
The picture grows clearer when we look at the number of
professor--student
collaborations. Here, relatively few of the UCSC faculty members have
nonempty sets of collaborators with their students.
Only Geoffrey K. Pullum and Jorge Hankamer have done such work
(Pullum with
Chris Barker,
Christopher Potts, and
Rachel Walker;
Hankamer with
Line Mikkelsen).
At UMass, the majority of the faculty members have had collaborations
with their past or current students. Here we list just a few; again,
each edge links to a page that documents the collaboration or
collaborations.
FACULTY
STUDENTS
Bach
Robin Cooper, Deirdre Wheeler, George Horn
Frazier
Ana Arregui,
Luis Alonso-Ovalle,
Katy Carlson,
Juli Carter,
Maria Nella Carminati,
Pat Deevy,
Mike Dickey,
Mako Hirotani,
Janina Rado,
Amy Schafer,
Elisabeth Villalta, and many others
Johnson
Satoshi Tomioka
Kingston
Christine Bartels, Laura Walsh Dickey
Kratzer
Junko Shimoyama
McCarthy
Linda Lombardi, John Alderete, Jill Beckman, Laura Benua,
Amalia Gnanadesikan, Suzanne Urbanczyk
Partee
Michael Bennett, Ji-yung Kim, Paul Portner, Mats Rooth
Pater
Adam Werle, Anne-Michelle Tessier
Potts
[James Isaacs, who was an undergrad at UMass
before heading to UCSC for graduate school]
Roeper
Bart Hollebrandse (and more than fifty others)
Selkirk
Shigeto Kawahara, Tong Shen, Mariko Sugahara,
Taka Shinya, and Koichi Tateishi
Chris Potts admits that including him in this binary relation is a
questionable move, since James Isaacs was not a student at UMass
when the two collaborated. But he wanted to be in the chart, and he
holds considerable sway with the WHISC editors, all of whom he has
caught in a compromsing position at one time or another. But Potts
hopes to gain a legitimate spot on the graph before 2004 is out.
He says, "In the words of the mathematician
Paul Erdös,
'My mind is open'".
ERDÖS
NUMBERS FOR UMASS LINGUISTS
Paul Erdös was the most prolific collaborator that mathematics
has ever seen. At the time of his death, he had
507
coauthors.
He collaborated with some of them dozens of times. Many were prolific
coauthors themselves. As a result, so-called
Erdös numbering
is now part of the
"folklore
of mathematicians throughout the world". One is said
to have a finite Erdös number just in case one can reach
Erdös himself in a finite number of collaboration
steps.[NOTE]
The rules for what counts as a collaborator are
here.
Erdös had an Erdös number of 0. His immediate coauthors have
Erdös numbers of 1. And so forth. If there is no such chain from
you back to Erdös, then your Erdös number is infinite.
Which UMass linguists have finite Erdös numbers?
It is very hard to calculate these things, so the following
graph might be incomplete (the edge links lead to pages
documenting the collaboration):
Update (2007): Potts's Erdos number is now 4, via this path: Paul Erdös[0] → Zsolt Tuza[1] → András Kornai[2] → Geoffrey K. Pullum[3] → Christopher Potts[4]. As
you can see, this puts Pullum's number at 3.
Ivan Niven
Samuel Eilenberg
Marcel Schützenberger
Noam Chomsky
Alan J Hoffman
PAUL ERDÖS
Emmon Bach (4)
Barbara Partee (4)
Morris Halle
Joseph B. Kruskal
Kenneth Kunan
Stanley Peters
Joan Bresnan
John J. McCarthy (6)
Mark Liberman
Jon Barwise
Larry Moss
Jane Grimshaw
Alan Prince
Robin Cooper
a coauthor of Emmon
Bach's, hence another
source of Emmon's 4
David E. Johnson
Paul M. Postal
Ivan Sag
Geoffrey K. Pullum
Janet Fodor
Christopher Potts (6)
Lyn Frazier (6)
If we consider the union of this graph with the
faculty collaboration graph,
then we obtain Erdös numbers for Angelika Kratzer (5), Tom Roeper
(6), and
Lisa Selkirk (7). If we consider the union of all the collaboration
graphs
in this WHISC, we obtain finite Erdös numbers for a host of UMass
students
past and present, which must be an unusual situation for a Humanities
department.
UPDATE (Jan 1, 2004)
John McCarthy writes, "Lest anyone get a swelled head, it's worth noting that a horse has an Erdös number of only 3."
[Many thanks to John McCarthy, Barbara Partee, and
Ivan Sag for help in compiling this graph.]
NOTE: It might be that all linguists
are excluded from having Erdös numbers, on the grounds
that these numbers hold only where the authors along the
path collaborated on mathematics papers.
But we'll ignore this point, for the
following reasons:
(i.)
This is a boring interpretation of Erdös numbering;
and (ii.)
At least one prominent linguist
has argued for years that linguists are basically set-theorists.
The above collaboration graphs are of course not the only ones,
and perhaps they are not even the most important ones. Which UMass
faculty members have collaborated with researchers in other UMass
departments? (Frazier, Roeper,...) Which have done interdisciplinary
work with nonlinguists at other universities? (McCarthy, Woolford,...)
These graphs would be large and difficult to compile. This is in itself
a testament to the interdisciplinary nature of linguistics, especially
as it is practiced at UMass.
