A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.
Journal of Personality and Social Psychology
1997,
Vol.72, No. 4, 863-871Copyright 1997 by the American Psychological Association, Inc.
0022-3514/97/$3.00
Linguistic Predictors of Adaptive Bereavement
James W. Pennebaker
Southern Methodist UniversityTracy J. Mayne
University of California, San Francisco
Martha E. Francis
Southern Methodist University
The words people use in disclosing a trauma were hypothesized to predict improvements in mental
and physical health in 2 studies. The first study reanalyzed data from 6 previous experiments in
which language variables served as predictors of health. Results from 177 participants in previous
writing studies showed that increased use of words associated with insightful and causal thinking
was linked to improved physical but not mental health. Higher use of positive relative to negative
emotion words was also associated with better health. An empirical measure that was derived from
these data correlated with subsequent distress ratings. The second study tested these models on
interview transcripts of 30 men who had lost their partners to AIDS. Cognitive change and empirical
models predicted postbereavement distress at 1 year. Implications of using computer-based text
analyses in the study of narratives are discussed.
Traumatic experiences, such as the death of a primary partner,
can affect psychological functioning on many levels. It can in-
fluence how people think about themselves, their past and future,
their relationships with others, and the meaning of life. Talking
with others about traumatic events is an important way in which
many people confront and resolve these issues (cf. Rime, 1995).
What people say about loss not only reflects their psychological
and emotional state, but may also aid them in coping with the
event.
The goal of this research was to examine the words people
use to discuss the recent death of a loved one and to predict
their subsequent physical and mental health. In the first of two
studies, writing samples from participants in six existing writing
studies were computer analyzed in an attempt to predict physical
health and psychological well-being. From these data, several
competing models were developed to specify exactly which as-
pects of language were associated with each of several out-
comes. In the second study, the models developed in Study 1
were used to predict the psychological functioning of 30 men
James W. Pennebaker and Martha E. Francis, Department of Psychol-
ogy, Southern Methodist University; Tracy J. Mayne, Center for AIDS
Prevention Studies. University of California, San Francisco.
This is one of four studies analyzing from different theoretical per-
spectives the bereavement narratives of 30 men whose partners had died
of AIDS. The data were from a longitudinal study: the University of
California, San Francisco, Coping Project.
Preparation of this article was made possible by National Science
Foundation Grant SBR-9411674 and National Institutes of Health Grants
MH52391 and MH440045. We are indebted to Diane Berry, Susan Folk-
man, and Tom Trabasso for their comments on earlier versions of this
article.
Correspondence concerning this article should be addressed to James
W. Pennebaker, Department of Psychology, Southern Methodist Univer-
sity, Dallas, Texas 75275. Electronic mail may be sent via the Internet
to pennebak@mail.smu.edu.
whose primary partners had died on the basis of two interviews,
performed 2 and 4 weeks after the deaths.
There is mounting evidence that the disclosure of emotional
experiences can have positive effects. In a series of laboratory
studies, individuals have been asked to write about traumatic or
other emotional experiences for 3-5 consecutive days, for 15-
30 min per day. Participants who have written about traumatic
events have shown improved health compared to control partici-
pants who have written about superficial topics. Those who have
written about traumatic events have been less likely to visit
physicians because of illness in the subsequent 2-14 months
(Pennebaker & Beall, 1986; Pennebaker, Kiecolt-Glaser, & Gla-
ser, 1988; Greenberg & Stone, 1992; Pennebaker, Colder, &
Sharp, 1990; Pennebaker & Francis, 1996; Krantz & Pennebaker,
1995;
Greenberg, Wortman, & Stone, 1996; Richards, Penne-
baker, & Beal, 1995). Experimental participants who have writ-
ten about traumas have also demonstrated greater imrnunocom-
petence, as measured by heightened blastogenic responses to
mitogens (Pennebaker, Kiecolt-Glaser, & Glaser, 1988), re-
sponses to Epstein-Barr antibodies (Esterling, Antoni, Fletcher,
Margulies, & Schneiderman, 1994), and higher antibody re-
sponses to hepatitis B vaccinations (Petrie, Booth, Pennebaker,
Davison, & Thomas, 1995). Writing about trauma has also been
found to be associated with positive behavioral outcomes, in-
cluding faster reemployment among laid-off engineers (Spera,
Buhrfeind, & Pennebaker, 1994) and higher grades among col-
lege students (Pennebaker, Colder, & Sharp, 1990; Pennebaker &
Francis, 1996; Krantz & Pennebaker, 1995).
On the basis of these and several other writing studies, we
have developed a general theory of disclosure and language. The
original theory posited that confronting upsetting topics reduced
the constraints or inhibitions associated with not talking about
the events. This was based on the assumption that inhibition
causes or exacerbates stress-related disease processes (Penne-
baker, 1989). In recent years, our working theory has evolved
863
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
