THE SCIENTIFIC STUDY OF ADOLESCENT DEVELOPMENT

In the opening sentence of the preface to the first edition of his classic A History of Experimental
Psychology, Edwin G. Boring (1929) reminded readers that “psychology has
a long past, but only a short history” (p. ix), a remark he attributed to the pioneer of
memory research, Hermann Ebbinghaus. A similar statement may be made about the
study of adolescents and their development.
The first use of the term adolescence appeared in the 15th century. The term was a
derivative of the Latin word adolescere, which means to grow up or to grow into maturity
(Muuss, 1990). However, more than 1,500 years before this first explicit use of the
term both Plato and Aristotle proposed sequential demarcations of the life span, and
Aristotle in particular proposed stages of life that are not too dissimilar from sequences
that might be included in contemporary models of youth development. He described
three successive, 7-year periods (infancy, boyhood, and young manhood) prior to the
person’s attainment of full, adult maturity. About 2,000 years elapsed between these initial
philosophical discussions of adolescence and the emergence, within the 20th century,
of the scientific study of the second decade of life.
The history of the scientific study of adolescence has had two overlapping phases
and is, we believe, on the cusp of a third. The first phase, which lasted about 70 years,
was characterized by three sorts of Cartesian splits (see Overton, 1998) that created
false dichotomies that in turn limited the intellectual development of the field. With respect
to the first of these polarizations, “grand” models of adolescence that purportedly
pertained to all facets of behavior and development predominated (e.g., Erikson,
1959, 1968; Hall, 1904), but these theories were limited because they were either largely
all nature (e.g., genetic or maturational; e.g., Freud, 1969; Hall, 1904) or all nurture
(e.g., McCandless, 1961). Second, the major empirical studies of adolescence during
this period were not primarily theory-driven, hypothesis-testing investigations but were
atheoretical, descriptive studies; as such, theory and research were split into separate
enterprises (McCandless, 1970). Third, there was a split between scholars whose work
was focused on basic developmental processes and practitioners whose focus was on
community-based efforts to facilitate the healthy development of adolescents.
The second phase in the scientific study of adolescence arose in the early- to mid-
1970s as developmental scientists began to make use of research on adolescents in elucidating
developmental issues of interest across the entire life span (Petersen, 1988). At
the beginning of the 1970s, the study of adolescence, like the comedian Rodney Dangerfield,
“got no respect.” Gradually, however, research on adolescent development began
to emerge as a dominant force in developmental science. By the end of the 1970s
the study of adolescence had finally come of age.
To help place this turning point in the context of the actual lives of the scientists involved
in these events, it may be useful to note that the professional careers of the editors
of this Handbook began just as this transition was beginning to take place. Across
our own professional lifetimes, then, the editors of this volume have witnessed a sea
change in scholarly regard for the study of adolescent development. Among those
scholars whose own careers have begun more recently, the magnitude of this transformation
is probably hard to grasp. To those of us with gray hair, however, the change has
been nothing short of astounding. At the beginning of our careers, adolescent development
was a minor topic within developmental science, one that was of a level of importance
to merit only the publication of an occasional research article within prime
developmental journals or minimal representation on the program of major scientific
meetings. Now, three decades later, the study of adolescent development is a distinct
and major field within developmental science, one that plays a central role in informing,
and, through vibrant collaborations with scholars having other scientific specialties,
being informed by, other areas of focus.
The emergence of this second phase of the study of adolescence was predicated in
part on theoretical interest in healing the Cartesian splits (Overton, 1998) characteristic
of the first phase and, as such, in exploring and elaborating developmental models
that reject reductionist biological or environmental accounts of development and instead
focus on the fused levels of organization constituting the developmental system
and its multilayered context (e.g., Sameroff, 1983; Thelen & Smith, 1998). These developmental
systems models have provided a metatheory for adolescent developmental
research and have been associated with more midlevel (as opposed to grand) theories—
models that have been generated to account for person-environment relations within
selected domains of development.
Instances of such midlevel developmental systems theories are the stage-environment
fit model used to understand achievement in classroom settings (Eccles, Wigfield, &
Byrnes, 2003), the goodness of fit model used to understand the relation of temperamental
individuality in peer and family relations (Lerner, Anderson, Balsano, Dowling,
& Bobek, 2003), and models linking the developmental assets of youth and communities
in order to understand positive youth development (Benson, 1997; Damon, 1997).
For instance, Damon (1997; Damon & Gregory, 2003) forwarded a new vision and vocabulary
about adolescents that was based on their strengths and potential for positive
development. Damon explained that such potential could be instantiated by building
new youth-community relationships predicated on the creation of youth charters, agreements
that codified community-specific visions and action agendas for promoting positive
life experiences for adolescents.
Generally speaking, the study of adolescence in its second phase was characterized
by an interest in developmental plasticity, in diversity, and in the application of science
to real-world problems. This phase also was marked by the development and use of
2 The Scientific Study of Adolescent Development
more nuanced and powerful developmental methods aimed at providing sensitivity to
the collection and analysis of longitudinal data pertinent to the multiple levels.
More than a quarter century ago, Bronfenbrenner (1974) explained the importance
of a science of development that involved the full and bidirectional collaboration between
the producers and consumers of scientific knowledge. In turn, D. A. Hamburg
(1992; D. A. Hamburg & Takanishi, 1996) proposed that the quality of life of adolescents,
and their future contributions to civil society, could be enhanced through collaboration
among scholars, policy makers, and key social institutions, for instance,
community-based youth-serving organizations (e.g., 4-H, Boys and Girls Clubs, scouting),
schools, and the media. In our view, D. A. Hamburg’s (1992; D. A. Hamburg &
Takanishi, 1996) vision has been actualized. We are now at the cusp of the emergence
of a third phase in the history of the scientific study of adolescence, one that we hope
will be marked by the publication of this Handbook. This phase involves the emergence
of the field of adolescent development as an exemplar of the sort of developmental science
that can be used by policy makers and practitioners in order to advance civil society
and promote positive development (Lerner, Fisher, & Weinberg, 2000). The contributors
to this volume provide much evidence that the field of adolescence may be
entering a phase of its development wherein such a scientist–policy maker–practitioner
collaboration may be a central, organizing frame.
THE FIRST PHASE OF THE SCIENTIFIC STUDY OF ADOLESCENCE
In 1904 G. Stanley Hall, with the publication of his two-volume work Adolescence, initiated
the scientific study of adolescence. He launched the field as one steeped in a split
and nativist view of development, one that was and linked to a biologically based,
deficit view of adolescence.
Fancying himself as the “Darwin of the mind” (White, 1968), Hall sought to translate
the ideas of Ernst Haeckel (e.g., 1868, 1891), an early contributor to embryology,
into a theory of life span human development. Haeckel advanced the idea of recapitulation:
The adult stages of the ancestors comprising a species’ evolutionary (phylogenetic)
history were repeated in compressed form as the embryonic stages of the organism’s
ontogeny. Hall extended Haeckel’s idea of recapitulation beyond the prenatal
period in order to fashion a theory of human behavioral development. To Hall, adolescence
represented a phylogenetic period when human ancestors went from being
beastlike to being civilized. Hall (1904) saw adolescence as a period of storm and stress,
as a time of universal and inevitable upheaval.
Although other scholars of this period (e.g., Thorndike, 1904) quickly rejected Hall’s
recapitulationism on both empirical and methodological grounds (e.g., see Lerner,
2002, for a discussion), other theorists of adolescent development used a conceptual
lens comparable to Hall’s, at least insofar as his biological reductionism and his deficit
view of adolescence were concerned. Anna Freud (1969), for instance, saw adolescence
as a biologically based and universal developmental disturbance. Erik Erikson (1950,
1959) viewed the period as one in which an inherited maturational ground plan resulted
in the inescapable psychosocial crisis of identity versus role confusion. Even when the-
The First Phase of the Scientific Study of Adolescence 3
orists rejected the nature-based ideas of psychoanalysts or neopsychoanalysts, they
proposed nurture-oriented ideas to explain the same problems of developmental disturbance
and crisis. For example, McCandless (1961, 1970) presented a social-learning,
drive-reduction theory to account for the developmental phenomena of adolescence
(e.g., regarding sex differences in identity development) that Erikson (1959) interpreted
as being associated with maturation (see Lerner & Spanier, 1980, for a discussion).
Although the developmental theory of cognition proposed by Piaget (1960, 1969,
1970, 1972) involved a more integrative view of nature and nurture than did these other
models, the predominant focus of his ideas was on the emergence of formal logical
structures and not on the adolescent period per se. The absence of concern in Piaget’s
theory with the broader array of biological, emotional, personality, social, and societal
concerns that had engaged other theorists’ discussions of adolescence did not stop a
relatively minor and historically transitory interest in Piaget’s ideas as a frame for empirical
understanding of the adolescent period (Steinberg & Morris, 2001). However, as
Steinberg and Morris explained, only a short while after this period of heightened interest
in using the onset of formal operations as an explanation for everything adolescent,
the influence of Piaget’s theory on mainstream empirical work in the study of adolescence
would become as modest as that associated with the other grand theories of
the period, such as those authored by Erikson or McCandless.
The divergence between the so-called grand theories of the adolescent period and
the range of research about adolescence that would come to characterize the field at the
end of the 20th century actually existed for much of the first phase of the field’s development.
The classic studies of adolescence conducted between 1950 and 1980 were not
investigations derived from the theories of Hall, Anna Freud, McCandless, Piaget, or
even Erikson (work associated with the ideas of Marcia, 1980, notwithstanding). Instead,
this research was directed to describing (note, not explaining; McCandless, 1970;
Petersen, 1988) patterns of covariation among pubertal timing, personal adjustment,
and relationships with peers and parents (e.g., Jones & Bayley, 1950; Mussen & Jones,
1957), both within and across cultural settings (e.g., Mussen & Bouterline Young, 1964);
the diversity in trajectories of psychological development across adolescence (e.g.,
Bandura, 1964; Block, 1971; Douvan & Adelson, 1966; Offer, 1969); and the influence
of history or temporality (i.e., as operationalized by time of testing- or cohort-related
variation) on personality development, achievement, and family relations (e.g., Elder,
1974; Nesselroade & Baltes, 1974). Petersen (1988, p. 584) described the quality of the
classic empirical work on adolescence by noting that most “research fell into one of two
categories: (a) studies on behavioral or psychological processes that happened to use
adolescent subjects, or (b) descriptive accounts of particular groups of adolescents,
such as high school students or delinquents.”
Despite its separation from the grand theories of adolescence that dominated the
field during its first phase of scientific development, this body of early research, as well
as the subsequent scholarship it elicited (e.g., see reviews by Lerner & Galambos,
1998; Petersen, 1988; Steinberg & Morris, 2001), made several important contributions
to shaping the specific character of the scientific study of adolescence between the
early-1980s and late-1990s. As elaborated later, this character involved the longitudinal
study of individual-context relations among diverse groups of youth and the use of such
scholarship for purposes of both elucidating basic developmental processes and apply-
4 The Scientific Study of Adolescent Development
ing developmental science to promote positive youth development (B. Hamburg, 1974;
Lerner, 2002).
These contributions also advanced the study of adolescence because scholarship
about the second decade of life acted synergistically with broader scholarly activity
within developmental science pertinent to the theoretical, methodological, and applied
features of the study of human development across the life span. For instance, a classic
paper by B. Hamburg (1974) did much to provide the foundation for this integration,
in that it made a compelling case for viewing the early adolescent period as a distinct
period of the life course and one that provided an exemplary ontogenetic window for
understanding key person-context processes involved in coping and adaptation. Based
on such evidence, Petersen (1988, p. 584) noted,
Basic theoretical and empirical advances in several areas have permitted the advance of research
on adolescence. Some areas of behavioral science from which adolescence researchers
have drawn are life-span developmental psychology, life-course sociology, social
support, stress and coping, and cognitive development; important contributing areas in
the biomedical sciences include endocrinology and adolescent medicine. The recent maturation
to adolescence of subjects in major longitudinal studies . . . has also contributed
to the topic’s empirical knowledge base.
The emergence of the relationship between the specific study of adolescence and more
general scholarship about the overall course of human development provided the bridge
to the second phase in the study of adolescent development. Indeed, about a decade after
this second phase had begun, Petersen (1988, p. 601) predicted, “Current research
on adolescence will not only aid scientific understanding of this particular phase of life,
it also may illuminate development more generally.” Future events were consistent with
Petersen’s prognostication.
THE SECOND PHASE OF THE SCIENTIFIC STUDY OF ADOLESCENCE
From the late 1970s through this writing the adolescent period has come to be regarded
as an ideal natural ontogenetic laboratory for studying key theoretical and methodological
issues in developmental science (Lerner, 2002; Steinberg & Morris, 2001). There are
several reasons for the special salience of the study of adolescent development to understanding
the broader course of life span development. First, although the prenatal
and infant period exceeds adolescence as an ontogenetic stage of rapid physical and
physiological growth, the years from approximately 10 to 20 not only include the considerable
physical and physiological changes associated with puberty but also mark a
time when the interdependency of biology and context in human development is readily
apparent (Susman & Rogol, this volume). Second, as compared to infants, the cognizing,
goal-setting, and relatively autonomous adolescent can, through reciprocal relations
with his or her ecology, serve as an active influence on his or her own development,
and the study of adolescence can inform these sorts of processes more generally
(Lerner, 2002). Third, the multiple individual and contextual transitions into, throughout,
and out of this period, involving the major institutions of society (family, peers,
The Second Phase of the Scientific Study of Adolescence 5
schools, and the workplace), engage scholars interested in broader as well as individual
levels of organization and also provide a rich opportunity for understanding the nature
of multilevel systemic change. Finally, there was also a practical reason for the growing
importance of adolescence in the broader field of developmental science: As noted by
Steinberg and Morris (2001), the longitudinal samples of many developmental scientists
who had been studying infancy or childhood had aged into adolescence. Applied
developmental scientists were also drawn to the study of adolescents, not just because
of the historically unprecedented sets of challenges to the healthy development of adolescents
that arose during the latter decades of the 20th century (Dryfoos, 1990; Lerner,
1995) but also because interest in age groups other than adolescents nevertheless frequently
involved this age group (e.g., interest in infants often entailed the study of
teenage mothers, and interest in middle and old age frequently entailed the study of the
“middle generation squeeze,” wherein the adult children of aged parents cared for their
own parents while simultaneously raising their own adolescent children).
The Emerging Structure of the Field of Adolescent Development
This scholarly activity at the close of the 1970s was both a product and a producer of a
burgeoning network of scholars from multiple disciplines. In 1981 the late Herschel
Thornburg launched a series of biennial meetings (called the Conference on Adolescent
Research) at the University of Arizona. During these meetings (which occurred
also in 1983 and 1985), the idea for a new scholarly society, the Society for Research on
Adolescence (SRA), was born. The first meeting of SRA was held in Madison, Wisconsin,
in 1986, and Thornburg was elected the first president of the organization. Across
the next two decades, with biennial conventions in Alexandria, Virginia (1988), Atlanta
(1990), Washington (1992), San Diego (1994), Boston (1996), again in San Diego (1998),
Chicago (2000), New Orleans (2002), and Baltimore (2004), and through the leadership
of the SRA presidents who succeeded Thornburg—John P. Hill, Anne C. Petersen,
E.Mavis Hetherington, Sanford M. Dornbusch, Jeanne Brooks-Gunn, Stuart T. Hauser,
Laurence Steinberg, W. Andrew Collins, Jacquelynne Eccles, and Elizabeth Susman—
the organization and the field it represented flourished. Between 1986 and 2002, attendance
at SRA biennial meetings more than quadrupled. The SRA launched its own
scholarly journal in 1991, the Journal of Research on Adolescence (Lerner, 1991); grew
from approximately 400 members in 1986 to more than 1,200 members in 2002; and
attracted disciplinary representation from scholars and practitioners with expertise
in psychology, sociology, education, family studies, social work, medicine, psychiatry,
criminology, and nursing.
Impetus to this growth in scholarly interest in the study of adolescence also was stimulated
by the publication in 1980 of the first handbook for the field. Edited by Joseph
Adelson (1980), the Handbook of Adolescent Psychology was published as part of the
Wiley series on personality processes. The volume reflected the emerging multidisciplinary
interest in the field (with chapters discussing levels of organization ranging from
biology through history, including an interesting historical chapter on youth movements),
the growing interest in systems models of adolescent development (e.g., in the
chapters by Elder, 1980, and by Petersen & Taylor, 1980), the importance of longitudi-
6 The Scientific Study of Adolescent Development
nal methodology (Livson & Peskin, 1980), and the increasing interest in diversity (i.e.,
there was a five-chapter section titled “Variations in Adolescence”). It is important to
note that through several chapters pertinent to the problems of adolescence there was
still ample representation in the volume of the deficit view of adolescence. Nevertheless,
the 1980 Handbook included information pertinent to normative development and developmental
plasticity, and several chapters discussed the positive individual and social
features of youth development.
The publication of a handbook, the organization of a successful scholarly society,
and the initiation of that society’s scholarly journal all underscored the growing interest
in and the scientific maturity of research on adolescent development. This intellectual
milieu and the scholarly opportunities it provided attracted a broad range of
scholars to the field, some for reasons that had little to do with adolescence per se, but
others because they came to see themselves as experts on the second decade of life. By
the mid-1980s a growing cadre of scientists would identify themselves as adolescent
developmentalists.
The Study of Adolescence as a Sample Case for Understanding Plasticity and
Diversity in Development
Scholars interested primarily in the instantiation of developmental processes within
other periods of the life span (e.g., infancy, Easterbrooks & Graham, 1999; adult development
and aging, Brim, 1966; Nesselroade & Baltes, 1974) or in disciplines other
than developmental psychology (e.g., life course sociology; Burton, 1990; Elder, 1974,
1980) became adolescent developmentalists as well. This attraction inheres in the
window that the period provides to understanding how development at any point
across the life span involves the relations of diverse and active individuals and diverse,
active, and multitiered ecologies (Bronfenbrenner, 1979; Bronfenbrenner & Morris,
1998; Lerner, 2002).
As suggested by Steinberg and Morris (2001), the scientific concern that arguably
was most significant in transforming the field of adolescent development beyond a focus
on this single developmental period into an exemplar for understanding the breadth
of the human life span was the emerging focus within developmental science on the
ecology of human development (e.g., Bronfenbrenner, 1979, 2001; Bronfenbrenner &
Morris, 1998). The integrated designed and natural ecology was of interest because its
study was regarded as holding the key to (a) understanding the system of relations between
individuals and contexts that is at the core of the study of human development and
(b) providing evidence that theories about the character of interacting developmental
system (e.g., Collins, Maccoby, Steinberg, Hetherington, & Bornstein, 2000; Gottlieb,
1997, 1998; Horowitz, 2000; Thelen & Smith, 1998) are more useful in accounting for
the variance in human ontogeny than are theories whose grounding is either exclusively
in nature (e.g., behavioral genetic or sociobiological; e.g., Plomin, 2000; Rowe, 1994;
Rushton, 2000) or exclusively in nurture (e.g., social learning or functional analysis;
Gewirtz & Stingle, 1968; McCandless, 1970).
A second set of broader issues that engaged developmental science in the study of
adolescence pertained to understanding the bases, parameters, and limits of the plas-
The Second Phase of the Scientific Study of Adolescence 7
ticity of human development. As implied earlier, this plasticity legitimated an optimistic
view about the potential for interventions into the course of life to enhance human development,
encouraged growth in scientific activity in the application of developmental
science to improve life outcomes, and gave impetus to the idea that positive development
could be promoted among all people (Lerner, Fisher, & Weinberg, 2000). Moreover,
plasticity meant that the particular instances of human development found within
a given sample or period of time were not necessarily representative of the diversity of
development that might potentially be observed under different conditions.
Third, developmentalists pursuing an interest in the developmental system and the
plasticity in ontogenetic change that it promoted recognized the need to develop and
deploy methods that could simultaneously study changes in (at least a subset of ) the
multiple levels of organization involved in the development of diverse individuals and
contexts. Accordingly, multivariate longitudinal designs were promoted as key to the
study of the relatively plastic developmental system, as were the development of empirical
tools, such as change-sensitive measures, sophisticated data analysis techniques,
and strategies such as triangulation of observations within and across both quantitative
and qualitative domains of inquiry.