ACKNOWLEDGMENTSI wish to express my gratitude to Vanderbilt University for
giving me the opportunity to spend much of 1986 in research
on the present topic, and to the American School of Classical
Studies in Athens for its hospitality and assistance in Septem-
ber of that year. For technical help I thank Bill Longwell, re-
sourceful director of Vanderbilt's microcomputer lab, and
Sharon Hardy. Other individuals to whom I am indebted in-
clude my colleague Alice Harris, who gave me timely tips on
some linguistic matters, and Emmett Bennett, at the Univer-
sity of Wisconsin, who explained for me what could and could
not be inferred from the agrimi tablets at Knossos. On a wide
range of topics I profited from the expertise of James Muhly
and William Wyatt, to both of whom I am thoroughly grateful
for saving me from error and for encouraging me along the
right paths. For errors that remain I am of course entirely re-
sponsible.
For permission to publish seven line drawings by Jaap Morel,
I thank the Allard Pierson Museum and the publishing house
of E. J. Brill, and I am indebted personally to Mr. Morel and
Dr. Mary Littauer and J. H. Crouwel. To Eve Pearson I am
grateful for her superb work, prompt and meticulous, in edit-
ing the manuscript. Finally, I thank Princeton University
Press and especially Ms. Joanna Hitchcock for transforming a
manuscript into a book.
IX