Windsor v. The United States Of America
REPLY AFFIDAVIT of Michael Lamb, Ph.D. in Support re: 28 MOTION for Summary Judgment.. Document filed by Edith Schlain Windsor. (Kaplan, Roberta)
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
EDITH SCHLAIN WINDSOR, in her
capacity as Executor of the estate of THEA
10 Civ. 8435 (BSJ) (JCF)
THE UNITED STATES OF AMERICA,
SUPPLEMENTAL EXPERT AFFIDAVIT OF MICHAEL LAMB, PH.D.
I, Michael Lamb, Ph.D., hereby depose and say as follows:
I am Professor of Psychology in the Social Sciences and Head of the Division
of Social and Developmental Psychology in the Faculty of Social, Human and Political
Science at the University of Cambridge.
On May 19, 2011,1 submitted my expert affidavit in this matter, which set
forth my relevant background and experience, and attached my curriculum vitae and a list of
my publications from the last 10 years, at Exhibit A.
That affidavit set forth the principal opinion that I am offering in this case:
that children and adolescents raised by same-sex parents are as likely to be well-adjusted as
children raised by heterosexual parents, including "biological" parents.
I have read the relevant portions of the Rule 56.1 statement and memorandum
of law filed by the Bipartisan Legal Advisory Group of the House of Representatives
("BLAG") in opposition to plaintiffs motion for summary judgment. More specifically,
various statements on pages 23 and 24 of BLAG's memorandum suggest that the research on
the psychological adjustment of children with gay and lesbian parents is not valid or reliable
because it has "serious flaws." I submit this affidavit in order to respond to this statement,
with which I strongly disagree.
There is No Basis for Dismissal of the Studies Referenced in My Affidavit as "Flawed".
First, BLAG asserts that studies referenced in my affidavit "comparing gay or
lesbian parents to heterosexual parents have serious flaws," and as supposed support for this
position, quote from three articles discussed at my deposition. (BLAG Opp. Br. at 23-24.)
This assertion is both misleading and false. When these quotations from other
scholars' articles were presented to me at my deposition in this case, I explained how they
had been taken out of context and that the scientific research on gay parent families is robust,
meets accepted rigorous standards for research in the field, and supports the central
conclusion provided in my prior affidavit and reiterated here: that children with gay and
lesbian parents are just as likely to be well-adjusted as those with heterosexual parents.
Specifically, in response to the first quotation from a report I had cited
["Studies of children raised by same-sex parents have almost exclusively focused on families
headed by lesbian mothers rather than gay fathers" (BLAG Opp. Br. at 23)], I explained at
[I]t's a fact that there have been more studies that are focused on the
adjustment of children raised by lesbians rather than by gay parents. It's a fact
that results of studies that are focused on the children in both of those contexts
are similar to one another. And it's a fact that in both contexts one finds that
the adjustment of children is affected not by the sexual orientation or by the
family structure but by the family process variables that we talked about
earlier this morning.
(Dep. Tr. at 16:6-11. f For example, Fair, Forsell and Patterson (2010) Parenting and child
development in adoptive families: Does parental sexual orientation matter? Applied
There are fewer studies of children raised by gay fathers than lesbian mothers largely
because there have been fewer gay men than lesbians raising children in their own
households presumably because, among other things, it is easier for lesbians to bring
biological children into their families. Indeed, only quite recently have significant numbers
of gay men begun to do so, too.
Developmental Science, 14(3), 164-178 (referenced in my initial affidavit) studied a sample
of children raised since infancy in families headed by gay male couples. The results were the
same as in the lesbian family research—child outcomes did not vary depending on the
parents' sexual orientation.
In other words, there is sufficient, indeed overwhelming, evidence that the
adjustment of children is not affected by their parents' sexual orientation, and the fact that
more studies have focused on children raised by lesbians, rather than by gay men, does
nothing to undercut that conclusion.
In response to BLAG's second quotation—"We still have relatively few
studies of adolescent offspring of lesbian or gay parents however, and some have advised
caution when generalizing the results of research conducted with young children to
adolescents" (BLAG Opp. Br. at 23)—I explained at my deposition as follows:
There are fewer studies of adolescents than there are of younger children. I
think that this statement here is part of a way of underscoring the importance
of this research. And it is important research. But there are several other
studies that have looked at adolescent offspring living with same-sex parents.
Does the fact that there are, I think you said, fewer studies on
adolescents counsel us to be cautious about drawing conclusions about
adolescents who are raised by same-sex parents?
Well, I don't think it does because the results of those fewer studies are
consistent with the results of other research that looks at children's adjustment
and other research that looks at adolescents and the factors that are associated
with their adjustment. Again, I think what's important to underscore is how
important it is to look at any set of findings in context. In that context, the fact
that studies like this show that children being raised by same-sex parents are
as likely to be well adjusted as children raised by ~ sorry - as adolescents
raised by heterosexual parents and that when one looks at the correlates of
better or worse adjustment, that it's the same factors regardless of sexual
orientation. It is the convergence between the findings and the broader body
of literature that is really the key thing we want to look at.
(Dep. Tr. at 82:15-83:21.)
As I explained, although there is less research on adolescents than on younger
children, there have been several studies involving adolescents and they have uniformly
reported positive outcomes on the part of adolescents raised by gay parents. Further, the
correlates of positive adolescent adjustment are the same regardless of the parents' sexual
The third quotation presented to me at my deposition and used by BLAG was
the following statement by Lawrence Kurdek:
Future research on gay and lesbian couples needs to address several key
issues. One is sampling. Because most studies have used convenience samples
of mostly white and well-educated partners, the extent to which findings
generalize to the larger population of gay and lesbian couples is unknown....
Most studies on gay and lesbian couples have used self-report surveys. Future
work could address some of the biases associated with self-report data . . . .
(BLAG Opp. Br. at 23-24.) When presented with this quote at my deposition, I explained as
Well, he's certainly correct in noting these issues in the literature on gay and
lesbian couples, particularly research of the sort that he has done, which has
painstakingly, and I think very usefully, shown that the dynamics of
relationships in gay and lesbian couples are characterized by the same
dimensions as those in heterosexual families and that clearly elaborating on
that and doing more research may be helpful for those who are interested in
further understanding couple dynamics. I do want to underscore that this, his
research, is focused on gay and lesbian couples, mostly couples without
children, and that these studies don't look at the relationship between the
couple variables and the children's adjustment. They are nevertheless very
useful because they do show that the research on the dynamics of those
couples is subject to and has the same sorts of correlates and variables as do
heterosexual couples, both those that are married as well as those who are cohabiting.
(Dep. Tr. at 85:10-86:7.)
In other words, researchers typically identify areas for future research in their
studies. This does not mean, as BLAG suggests, that the studies are invalid or unreliable. As
I explained, Kurdek's research on same-sex couples' relationships is both rigorous and
reliable. It is also important to note that he was not discussing research on gay parents, as
BLAG incorrectly suggests.
Contrary to BLAG's misleading presentation, therefore, I have never
suggested that the scientific research on gay parent families is flawed or deficient; rather, I
made it clear that the research is not only robust, but reliably shows equally good outcomes
for children of gay and heterosexual parents. None of the quotations that BLAG cites
undercuts the validity of the research. BLAG fundamentally mischaracterizes the scientific
evidence, more fully summarized in my affidavit to which an extended bibliography was
The Sources BLAG Cites Do Not Reflect the Science
In further attempts to support its position that the gay parenting research
should be dismissed because of methodological flaws, BLAG also made reference to a
decision of the Eleventh Circuit Court of Appeals in the Lofton case (addressing a challenge
to a Florida law banning adoption by gay people) and two articles, one written by Anne
Hulbert, a writer, and one by George Dent, a law professor. Based on these writings, BLAG
claims that, "Numerous studies have pointed to methodological flaws in those studies
comparing heterosexual and homosexual parents." (BLAG Opp. Br. at 24.)
That is not the case. The Hulbert and Dent materials are non-scientific
sources and, thus, are fundamentally unreliable. As an initial matter, neither Anne Hulbert
nor George Dent have professional expertise with respect to child development. In addition,
their articles were not published in peer-reviewed scientific journals. Indeed, the three page
Some of the "authorities" that Dent cites in his article were authored by individuals who
have been discredited as biased with respect to the research on homosexuality. For example,
Dent cites Paul Cameron, the founder of an anti-gay advocacy organization called the Family
Research Institute, who reportedly had his membership in the American Psychological
Association revoked. He also cites George Rekers and Walter Schumm. Two courts have
found Rekers' testimony about gay parents to be biased. As a Florida court explained:
"[His] testimony was far from a neutral and unbiased recitation of the relevant scientific
evidence. Dr. Rekers' beliefs are motivated by his strong ideological and theological
convictions that are not consistent with science. Based on his testimony and demeanor at
trial, the court cannot consider his testimony to be credible nor worthy of forming the basis of
public policy." In re Adoption of Doe, 2008 WL 5006172, at *12 (Fla. Cir. Ct. Nov. 25,
article by Hulbert, which was published in a popular on-line magazine, www.slate.com,
contains an errata footnote acknowledging her earlier misunderstanding of statistics presented
in a study.
The Eleventh Circuit's characterization of the gay parenting research also does
not match reality. The sources cited by the Lofton court purporting to show that the research
is flawed include: (1) a 1995 commentary by Baumrind calling for the very kinds of studies
that have been published in the years since; (2) a booklet written by Lemer and Nagai that
was published by the Marriage Law Project, an advocacy organization that opposes marriage
for same-sex couples; and (3) a review article written by Stacey and Biblarz that, contrary to
the Court's characterization, deems the research sufficiently strong and reliable to draw the
Because every relevant study to date shows that parental sexual orientation,
per se, has no measurable effect on the quality of parent-child relationships or
on children's mental health or social adjustment, there is no evidentiary basis
for considering parental sexual orientation in decisions about children's best
Contrary to the conclusion of the Lofton court, the research on gay parent
families is a robust body of research that meets the rigorous methodological standards
2008); see also Howard v. Child Welfare Agency Review Bd, 2004 WL 3154530 (Ark. Cir.
2004), aff d, 238 S.W.3d 1 (Ark. 2006)CTt was apparent from both Dr. Rekers' testimony
and attitude on the stand that he was there primarily to promote his own personal ideology. If
the furtherance of such ideology meant providing the court with only partial information or
selectively analyzing study results that was acceptable to Dr. Rekers. . . . Dr. Rekers'
willingness to prioritize his personal beliefs over his function as an expert provider of fact
rendered his testimony extremely suspect and of little, if any, assistance to the court in
resolving the difficult issues presented by this case."). As for Walter Schumm, the same
Florida court that criticized Rekers concluded that Schumm "integrates his religious and
ideological beliefs into his research." Id.
BLAG also cites in its motion to dismiss an article by Lynn Wardle as purported support
for its argument that children raised by same-sex parents miss the benefits of being raised by
men and women. (BLAG Mem. in Support of Mot. to Dismiss at 43.) Like Dent, Lynn
Wardle is a law professor with no professional expertise in child development and his article,
which was published in a law journal, is not a scientific publication.
The Hulbert article from slate.com similarly misinterprets Stacey and Biblarz.
demanded for publication in the leading academic journals. There is simply no basis on
which to dismiss this body of research as invalid or unreliable due to methodological
Of the two studies cited by the Lofton court to purportedly show worse
outcomes for children with gay parents, the first, a publication by Paul Cameron, is worthless
because, although it claims to show that children are more likely to be molested by gay
parents than heterosexual parents, it makes clear that the researchers did not even ask
respondents about the sexual orientation of their parents. See also n. 1, supra. Second, the
Stacey and Biblarz review discussed above did not conclude that children with gay parents
fared worse than children with heterosexual parents. In fact, it concluded that there is "no
evidentiary basis for considering parental sexual orientation in decisions about children's best
interest." As discussed more fully in my original affidavit, the research on children raised by
gay parents consistently shows that parental sexual orientation has no bearing on child
outcomes. There can be no reasonable doubt that this is a matter of scientific consensus.
Finally, BLAG attempts to refute my statement that "[t]here is no empirical
support for the notion that the presence of both male and female role models in the home
promotes children's adjustment or well being," by referencing studies cited in the case
Mzarry v. Bd ofEduc. of Chi., 251 F.3d 604, 607 (7th Cir. 2001). (BLAG Rule 56.1 Stmt.
Tf47.) As BLAG notes, those studies show only that so far as heterosexuals are concerned, the
evidence that marriage "provides a stable and nourishing framework for child-rearing .. .
In addition, while BLAG did not cite it in the Windsor case, in Pedersen, BLAG cited
Norval Glenn's criticism of sampling in research on gay/lesbian parenting: See BLAG's
opposition to summary judgment in Pedersen, pp. 31-32 (citing Glenn's article, "The
Struggle for Same-Sex Marriage"). For the same reasons discussed above, Glenn's
article is without merit. Although Glenn dismisses this body of research for using small
convenience samples, such sampling is appropriate and routinely used in psychological
research. Moreover, a number of the studies on gay parent families did use representative
refutes any claim that policies designed to promote marriage are irrational." Significantly,
the Mzarry court and these studies were not comparing families of heterosexual and gay
couples—they were simply comparing married and non-married heterosexual couples.
BLAG also cites a 1974 review by Biller extolling the positive impact that heterosexual
fathers can have on their children's development. The speculation nearly 40 years ago that
that children needed to have both male and female parents in order to be well-adjusted has
not been supported by the research. Research conducted since then has demonstrated that
both mothers and fathers are important to their children as parents, not as males and females,
and that the parents' genders do not affect children's adjustment.
Finally, the sources cited by BLAG are not only unscientific and do not
support their arguments, but were largely published prior to 2004. As discussed above and in
my original affidavit, the body of scientific research pre-dating 2004 fully supported my
opinion that child adjustment is not affected by the parents' sexual orientation. Since then,
however, there has been much additional research exploring the psychological adjustment of
children and adolescents raised by same-sex couples. (I cited forty-three academic articles
published after 2004 in Exhibit B appended to my initial affidavit.) This more recent research
only serves to underscore the established conclusion that child adjustment is not in any way
affected by parental sexual orientation.
In its motion to dismiss at 40-41, n. 11, BLAG says that "In 2004, Congress extensively
reviewed the evidence that children whose mother or father is absent are comparatively
worse off," and references an article by Barbara Dafoe Whitehead published in the
Atlantic magazine (http://www.theatlantic.eom/magazine/archive/l993/04/dan-quaylewas-right/7015/) and a report by the organization Child Trends
(http://www.childtrends.org/files/marriagerb602.pdf) as examples. These articles discuss
the research on family dissolution and single parenthood among heterosexual parents,
noting poorer average outcomes for children whose family lives are disrupted by divorce
and those who do not have supportive relationships with both of their parents. These
articles do not, as BLAG suggests, say anything to support the claim that parents' gender
affects children's adjustment. And they have no relevance to gay parent families other
than to suggest that children of gay parents would also be more likely to be well adjusted
if their parents' relationships remained intact.
The conclusion stated in my affidavit—that children with gay, lesbian, and
heterosexual parents are all as likely as one another to be well-adjusted because it is the
quality of parenting and parent-child relationships, the quality of the parents' relationships
with partners and other significant adults, and their social and economic resources, rather
than their family structure or type that determines children's outcomes—is well supported by
the empirical research literature, and is not undermined by the misleading assertions made by
BLAG in opposing plaintiffs motion for summary judgment.
I declare under penalty of perjury under the laws of the United States of America that
the foregoing is true and accurate.
Executed this [ o th day of September 2011 at Washington, DC.