What's in a Name? The Disparate Effects of Identifiability on Offenders and Victims of Sexual Harassment

• Published date: 01 December 2019
• Author: Daphna Lewinsohn‐Zamir,Netta Barak‐Corren
• DOI: http://doi.org/10.1111/jels.12233
• Date: 01 December 2019

Journal of Empirical Legal Studies

Volume 16, Issue 4, 955–1000, December 2019

What’s in a Name? The Disparate Effects of

Identifiability on Offenders and Victims of

Sexual Harassment

Netta Barak-Corren*and Daphna Lewinsohn-Zamir

Sexual harassment is undergoing an identification revolution, as more victims choose to

forego their anonymity and divulge their identity to the public. Research in social psychol-

ogy on the identifiability effect has found that identified victims typically generate more

empathy and support than unidentified ones. However, this research has been limited

largely to monetary donations or to unambiguous cases with uncontested facts; the schol-

arship has not examined the effects of varying the identifiability of both parties to a con-

flict. In three large-scale experiments with a representative population (total N= 3,988),

we found that in the context of sexual harassment, victims do not gain an identifiability

“premium”—whereas offenders do. Offenders identified by their first name only are reg-

arded as more credible and moral and less blameworthy and responsible for the event

than unidentified offenders, but the same does not apply to identified victims. Further-

more, when the offender is identified, fewer people perceive the case as involving sexual

harassment (Experiment 1), and support for taking measures against the offender

declines (Experiment 2). Finally, the identified offender premium exists for offenders of

both sexes, but the detrimental effect of identification on victims is moderated by the vic-

tim’s mode of identification. Specifically, identified female victims who stated willingness

to disclose their name publicly fared worse than those preferring that their name not be

revealed in public, and the difference between active and passive identification reversed

for male victims. The effect of identification mode is moderated by sexist beliefs

(Experiment 3). Our results have normative implications for the appropriate balance

between publicity and anonymity in various contexts, including social networks, the media,

and disciplinary and judicial tribunals.

*Address correspondence to Netta Barak-Corren, Faculty of Law, Hebrew University of Jerusalem, Mt. Scopus, Jeru-

salem 9190501, Israel; email: barakcorren@mail.huji.ac.il. Barak-Corren is Assistant Professor of Law, Faculty of

Law, Hebrew University of Jerusalem; Lewinsohn-Zamir is Dean and Louis Marshall Professor of Environmental

Law, Faculty of Law, Hebrew University of Jerusalem.

For their helpful comments and suggestions, we thank Ronen Avraham, Shoham Choshen-Hillel, Fiery Cush-

man, Yuval Feldman, Miri Gur-Arye, Dave Hoffman, Stephanie Holmes Didwania, Ori Katz, Tehila Kogut, Tami

Kricheli-Katz, Jeff Rachlinski, Ilana Ritov, Yoram Shachar, Amos Schurr, Doron Teichman, Eyal Zamir, and partici-

pants in the Harvard & MIT negotiations seminar, the Harvard Moral Psychology group, the 13th Annual Confer-

ence on Empirical Legal Studies (Michigan Law School), the 2018 Society for Judgment and Decision Making

Annual Meeting, the Work-in-Progress Workshop at Chicago Law School, the Law and Economics Workshop at

Tel-Aviv University, and faculty seminars at Northwestern Law School, the Hebrew University, and the Interdisci-

plinary Center Herzliya. Chagit Blass, Afik David, Aviel Gordis, and Itamar Granot provided excellent research

assistance. This research was supported by the Israel Science Foundation (Grant No. 128/16).

955

I. Introduction

At present, the fight against sexual harassment seems to be in full swing. What began with

the individual initiative of a few women—notably actresses such as Ashley Judd and Alyssa

Milano—who came forward with their personal stories, gained great momentum worldwide.

Extensive media coverageand the hashtag #MeToo used bymillions of people have brought

the issue of sexual misconduct to the forefront of public attention. As a result, powerful,

long-time offenders—like movie mogul Harvey Weinstein—were forced to resign from their

positions (for a short survey of the high-profile incidents giving rise to the #MeToo move-

ment, see Wexler et al. 2018). Indeed, Time magazine collectively chose the people who

broke the silence around this pervasive social problem as its 2017 “Person of the Year.”

1

Most of the persons depicted in the Time article appear with their full name and a large

(sometimes full-page) color photograph. Although the article features testimonies from low-

profile individuals as well, such as a dishwasher or a hotel housekeeper, the vast majority of

interviewees are successful or even famous professionals, including movie and media stars,

artists, engineers, politicians, university professors, and journalists. Interestingly, the two

women to maintain anonymity (and thus photographed from the back) belong to the for-

mer group: a hospital worker and an office assistant; the immigrant strawberry picker whose

photo (from the front) appears on thecover of the magazine used a pseudonym.

In general, information about alleged offenders and victims of sexual misconduct

comes in varying degrees of anonymity or identification from the perspective of its recipi-

ents.

2

Thus, in a newspaper report, a TV program, or an online blog, both parties may be

anonymous; both may be identified; or only onepartymaybeidentifiedwhiletheother

(either the victim or the offender) remains anonymous. Furthermore, anonymity can be

complete or incomplete; glimpses and hints of identity can be provided, for example, in the

form of the initials of the person’s name, a photo taken from behind, or a pixelated image.

Likewise, identification is a matter of degree and can include only the person’s first name or

also the surname, a clear photo of his or her face, bibliographical information, and so on.

This stateof affairs raises the question of whetheranonymity or identification matters.Is

our evaluation of an event influenced bywhether the people depicted in it are anonymous or

identified? Specifically, in the context of sexual misconduct, does this factor affect our judg-

ments about the credibility, blame, and morality of the accuser and the accused, or whether

sexual harassment in fact occurred? To the extent that such judgments are indeed influenced

1

See http://time.com/time-person-of-the-year-2017-silence-breakers/. At the same time, concerns have been raised

about lack of due process and disproportionate consequences for some of the people accused of sexual miscon-

duct. For discussion of this issue, see Wexler et al. (2018).

2

For the sake of simplicity, we hereafter use the terms “victim” and “offender” in a broad manner that encompasses

the person complaining about sexual harassment and the person accused of sexual harassment, respectively. These

terms should not be understood as necessarily implying that the alleged victim had been harassed by the alleged

offender. Indeed, our experimental vignettes examined people’s judgments about a sexual harassment case in

which each party presented a conflicting version of an event (and the vignettes themselves did not use these

terms). Finally, while the terms “offender” and “victim” might involve a connotation of criminal liability, in the pre-

sent context they equally apply to disciplinary and civil proceedings.

956 Barak-Corren & Lewinsohn-Zamir

by whether ornot the individuals involved are anonymous, this phenomenon bears potential

social and legal implications. For example, if even minimal identification—say, by first name

alone—increases or decreases a person’s credibility or blameworthiness in the eyes of others,

this, in turn, might affect the willingness of victims (most of whom are “ordinary” people

rather than celebrities) to complain about sexual harassment, as well as the treatment they

will receive in their surroundings. The importance of this inquiry is augmented by the fact

that allegations of sexual harassment are often dealt with in the public arena (Wexler et al.

2018) and do not necessarily culminate with a decision by a court or a disciplinary board.

Given that anonymity or identification is an inherent, prominent feature of any report

on sexual harassment, it is surprising that this issue has not yet been examined through the

lens of the psychological phenomenon labeled the identifiability effect. The identifiability effect

is the tendency of peopleto react more strongly toidentified individuals than to unidentified

ones. It was found, for example, that victims of natural disasters or individuals who are gravely

ill usuallyattract more generousdonations from the public when they are identifiedby name

and/or a photo, rather than when they are anonymous (Kogut & Ritov 2005b; Small et al.

2007). In this study, we test the existence and characteristics of the identifiability effect in a

heretoforeunexplored andsocially importantcontext—sexualharassment.

3

Our studyreports on the findingsof three originalexperiments thatexamined people’s

judgments regarding a sexual harassment case using representative samples of Israeli society.

We found that minimal and meaningless identification—by first name only—significantly

impacted the parties to the event. Generally speaking, such identification benefited the

offender: respondents viewed an identified wrongdoer (either male or female) as more credi-

ble and moral, and less blameworthy and responsible for the event (relative to the victim),

than an unidentified wrongdoer. Furthermore, when the offender was identified, respon-

dents were less inclined to regard the situation as one that involves sexual harassment. In con-

trast, identification was much less beneficial for the victim and,under certain circumstances,

could even worsen her or his position. In particular, a female victim fared worse when she

identified actively by stating her willingness to disclose her name publicly (rather than prefer-

ring thather name not be revealedin public; hereinafter “passiveidentification”).Active iden-

tification resulted in a perception of less credibility and morality, more blameworthiness and

responsibility for the event, less determinations that sexual harassment had in fact occurred,

and less support for taking measures against the offender. We examined various explanations

for this disparate effect of identification on offenders and victims, including respondents’

gender, their emotional reactions, and sexism.

4

Briefly,wefindthatthedifferencebetween

active and passive identification reverses for male victims (with passively identified males far-

ing worse than active ones) and is moderated by sexism.

3

For a general discussion of sexual harassment and the diverse legal and social issues involved, see, for example,

MacKinnon and Siegel (2004), Marshall (2005), and LeMoncheck and Sterba (2001).

4

We also examined a series of additional proposed mediators, including political affiliation, religious affiliation,

age, social norms perceptions, fears related to sexual harassment, and stereotypical perceptions regarding the char-

acteristics of the offender and the victim. For the sake of brevity, and since none of these factors moderated the

effect, we report these analyses in the Appendix.

Identifiability and Sexual Harassment 957