Content uploaded by Melodie Fox
Author content
All content in this area was uploaded by Melodie Fox on Jul 19, 2016
Content may be subject to copyright.
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
328
Prototype Theory:
An Alternative Concept Theory
for Categorizing Sex and Gender?
Melodie J. Fox
School of Information Studies, University of Wisconsin, Milwaukee, 3210 N. Maryland Ave.,
Bolton Hall #510, Milwaukee, WI 53211, USA, <msjoblad@uwm.edu>
Melodie J. Fox is a doctoral student at the University of Wisconsin-Milwaukee’s School of Information
Studies and is a member of the Information Organization Research Group there. She holds an MLIS
from the University of Wisconsin-Milwaukee and an MA in English from the University of Illinois at
Chicago. Her research interests involve the social consequences of the organization and representation
of information, with particular focus on gender and race.
Fox, Melodie J. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gen-
der? Knowledge Organization, 38(4), 328-334. 18 references.
ABSTRACT: Classical theories of classification and concepts, originating in ancient Greek logic, have
been criticized by classificationists, feminists, and scholars of marginalized groups because of the rigidity of conceptual bounda-
ries and hierarchical structure. Despite this criticism, the principles of classical theory still underlie major library classification
schemes. Rosch’s prototype theory, originating from cognitive psychology, uses Wittgenstein’s “family resemblance” as a basis
for conceptual definition. Rather than requiring all necessary and sufficient conditions, prototype theory requires possession of
some but not all common qualities for membership in a category. This paper explores prototype theory to determine whether it
captures the fluidity of gender to avoid essentialism and accommodate transgender and queer identities. Ultimately, prototype
theory constitutes a desirable conceptual framework for gender because it permits commonality without essentialism, difference
without eliminating similarity. However, the instability of prototypical definitions would be difficult to implement in a practical
environment and could still be manipulated to subordinate. Therefore, at best, prototype theory could complement more stable
concept theories by incorporating contextual difference.
1.0 Introduction
Classical theories of classification and concepts, ori-
ginating in ancient Greek logic, have been roundly
criticized by classificationists, feminists, and scholars
of marginalized groups because of the rigidity of con-
ceptual boundaries and hierarchical structure, rife
with implications of dominance and subordination
(Olson 2001, 2007). While classical theory works well
in simple situations, human experience is faced with
ambiguity, inconsistency and incompleteness of in-
formation, as the world is dynamic and full of over-
lap. Classical theory struggles to handle epistemo-
logical variation, especially with complex, socially-
influenced categories, such as groupings of people.
The options are either to erase difference, to fudge
over complexity, or to classify so closely that the
categories are tiny. Despite criticism, in LIS, classical
theory still underlies the major classification systems.
Perhaps the problem lies not in the categorization
of humans, but rather in the concept of a concept.
Concepts facilitate the ability to categorize and thus
understand and predict the material world, yet dis-
agreement exists on what formulates a concept. Most
scholarly work about concepts begins with the ac-
knowledgement that a satisfactory overall theory of
concepts does not exist and would be difficult to
conceive. It is not my purpose to come up with a
complete concept theory. Rather, I will analyze one of
the existing concept theories to determine how it
handles slippery and sensitive concepts for which ca-
tegorization has real consequences: sex and gender.
At a cognitive level, classification helps humans
function by reducing what we see into concepts, al-
lowing us to recognize like items, thus shortening
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
329
processing time as we move through our day. Often,
the shortcuts that concepts provide can blur signifi-
cant differences within a category, particularly when
classifying people. For those who fall in the border-
land between male and female or those who resist the
gendered behavior socially prescribed for their sex,
the mutually exclusive gender categories afforded by
classical theory can cause social and emotional conse-
quences. A broader cultural understanding of sex and
gender can influence systemic discourses whether in
legal, medical, social, educational, or information en-
vironments. This paper, focusing on the theoretical, is
the first stage of ongoing research exploring different
ways to categorize sex and gender.
Would the concepts of sex or gender change if the
concept of concepts changed? To investigate, I under-
take three tasks that Haslanger (2000, 33) calls con-
ceptual, descriptive and analytical inquiries. For the
conceptual inquiry, I will explore prototype theory as
an alternative or supplementary concept theory to de-
termine whether it would be a viable option for defin-
ing sex and gender in ways that reflect the diversity
that classical concept theory cannot capture. For the
descriptive inquiry, I will use Jacob Hale’s (1996, 290)
“Defining Characteristics of the Category Woman” as
an example of a gender prototype. And for the ana-
lytical approach, I will discuss the implications of a
different concept theory for sex and gender. At first
blush, prototype theory constitutes a desirable con-
ceptual framework for gender because it permits
commonality without essentialism, difference with-
out eliminating similarity. However, the instability of
prototypical definitions would cause difficulty in a
practical environment. Therefore, at best, prototype
theory could complement more stable concept theo-
ries by incorporating contextual difference.
2.0 Concept theory
The study of concepts dates back to ancient Greek
philosophy, and debates persist over the philosophi-
cal, psychological, metaphysical, linguistic, epistemo-
logical and cognitive implications of concepts (Mar-
golis and Lawrence, 2006). Generally speaking, con-
cepts are universals whose particulars reside together
in a category. Concepts do not have a one-to-one re-
lationship with language, as different terms can be
used to express the same concept. The structure of
concepts is determined by what irreducible condi-
tions the concept requires and how those parts are es-
tablished.
2.1 Classical concept theory
Classical concept theory relies on the classical logic of
Parmenides and Plato, further developed by Aristotle
in the fourth century BCE. Many scholars have cri-
tiqued classical theory’s focus on hierarchy, mutual
exclusivity, and the inherent essentialism of its con-
ceptual structure, all qualities inhospitable to classify-
ing people. Andrea Nye (1990) and Hope Olson
(1999, 2007) have tracked the evolution from Pla-
tonic divisions to Aristotelian classification from a
feminist perspective in detail, so this paper will only
provide a rudimentary overview to supply some sup-
porting information and a basis for comparison.
In classical concept theory, “a concept is a summary
representation of some sets of things in terms of con-
ditions that are singly necessary and jointly sufficient
for determining membership in that set” (Hjørland
2009, 1520-21). Parmenides first introduced divisions
between concepts based on difference, expressed as
sets of A/not A. All items in set A possess the condi-
tions or qualities outlined in the definition of the con-
cept. Although classical theory purports to name an
extant, orderly nature, the qualities or conditions se-
lected to determine difference generally are based on
the goal of the classification scheme (Olson 1999, 69).
The traits chosen might be arbitrary, chosen to sim-
plify the classification process, or for more specious
reasons, such as to deliberately classify an item or
group into a subordinate position.
Plato intended for impermeable walls to partition
concept sets. The divisions are governed by three laws,
leading to mutual exclusivity: the Law of Non-
contradiction: Nothing can be both A and Not-A; the
Law of Identity: Whatever is A is A; and the Law of
the Excluded Middle: Everything is either A or Not-A
(Olson 2007, 511). Mutual exclusivity requires that an
item cannot partially be a member of a set or a mem-
ber of two sets simultaneously; membership is all or
nothing. Not A is not the opposite of A, but rather
the absence of A. An item that does not possess all re-
quired qualities would have no value. No credit is
awarded for being “kind of like” the other items in the
set.
Classical theory’s rigidity does not allow for the
shifting nature of social categories and in fact intends
for conceptual definitions to be isolated from context
(Slaughter 1988, 9). Frye (2005, 48-49) insists that
“social categories are not sets, and thinking of them
as sets is disastrous.” She differentiates social catego-
ries from Plato’s strict divisions by writing (49, em-
phasis original):
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
330
However social categories work … they do it
without having boundaries fixed by necessary
and sufficient conditions and consequently
without having any absolute sameness as their
principle of coherence.
Campbell (2000, 127) speaking of gay and lesbian
communities through the framework of Sedgwick
suggests, “people are different, categories shift, and
labels are provisional. We are dealing with no mono-
lithic identity, no stable categories, and no consensus.”
Social sets, then, are unstable entities that contain a
variety of similar members, with shifting and porous
boundaries.
2.2 Prototype theory
Prototype theory provides a more hospitable envi-
ronment for social categories. Wittgenstein ([1988]
1953) designates “family resemblance” as the thread
that constructs the fabric of a concept. He acknowl-
edges the difference that occurs within categories
(66): “For if you look at them you will not see some-
thing common to all, but similarities, relationships,
and a whole series of them,” a “complicated network
of similarities overlapping and criss-crossing.” The
abstract conceptualizations provided through classical
concept theory are too vague to really discern the
meaning of a word. Instead, context is required to
recognize how the word is being used. Wittgenstein
calls the shifting of meaning “language games,” and
this idea helped shape prototype theory as developed
by Rosch.
Rosch’s (1999) prototype theory of concepts,
originating in cognitive psychology, also decides
membership in a category through possession of par-
ticular properties, but group membership does not
require possession of all qualities—just enough for
the fabric of the concept to hold together. Just as
Wittgenstein believed, context determines the mean-
ing of a word or what conditions are in place at a par-
ticular moment to define the concept. Qualities of a
concept are incidental, rather than essential. The
member probably will possess all features typical to
the group, but if it possesses some but not all, it still
retains membership in the group. Unlike a universal
scheme where concepts are defined the same every
time, concepts are defined “only in actual situations
in which they function as participating parts of the
situation rather than as either representations or as
mechanisms for identifying options” (61). Member-
ship in a category can be determined by resemblance
to an ideal exemplar or by possession of a sufficient
number of the typical features of the class. A proto-
type might be either a “paragon” or an “average”
member of the group, and the prototype differs be-
tween individuals and moments, making it contextu-
ally variant. A concept might also contain multiple
prototypes. The prototype anchors the ideational
content of the concept, but does not exclude variants,
nor does it deem difference as deviant.
2.3 Concept theory in KO
In knowledge organization, Hjørland (2009) de-
scribes concepts as products of specific disciplinary
domains and argues that differing knowledge organi-
zations systems can co-exist based on different con-
ceptualizations found in different domains. A classi-
fication scheme should be “linked to certain dis-
courses and interests,” acknowledging historical and
paradigmatic associations (1529). While Hjørland’s
view is appealing and could work well within special-
ized libraries, it has been criticized for building obsta-
cles for interdisciplinary research (Szostak 2010), and
its usefulness in universal settings, such as public li-
braries, merits further investigation.
Szostak (2004) proposes instead a universal
scheme for classifying science that breaks complex
concepts down into the most basic “constituent
parts,” upon which scientists can agree, therefore
making concepts accessible across disciplines.
Szostak’s view is contingent upon belief in universal
concepts, which can be incompatible with postmod-
ern epistemologies. Dahlberg (1995, 22) adheres to
the classical formula, where essential qualities are out-
lined in defining a concept. She calls her approach the
“analytical, referent-oriented” concept theory, where
the analysis must be based on an “item of reference.”
In her model of concept construction, “correct state-
ments” must be made about the item; however, “cor-
rect” statements may be difficult to verify unless,
again, a belief in universal truths exists. Bowker and
Star (1999) believe that humans use both classical
theory and prototype theory when classifying. Classi-
cal theory provides universality and stability, and pro-
totype theory accommodates the fuzziness of so-
cially-determined variance. They write that classifica-
tion “involve[s] politics, kinds both prototypical and
Aristotelian … and deletion of the practices in the
production of the final formal record” (66). Jacob
(2004, 537, 522) makes a similar distinction between
categorization, which is “flexible and creative,” and
classification, which is “artificial and arbitrary.”
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
331
3.0 The Concept of Woman
The current state of feminism resists classical concep-
tions of woman and has taken a tactic similar to Hjør-
land’s domain-centric approach, where each theory
produces an eristic concept of woman not always
compatible with those of other theories. In the disci-
pline of gender studies, gender has been defined as a
social construct, quirk of psychosocial development,
biological set, inconsequential human trait, perform-
ance, sociolinguistic activity, and self-perpetuating il-
lusion, among others. As Mikkola (2009, 560) legiti-
mately points out, questions about the definition of
“woman” are puzzling to most people: “Aren’t women
simply female humans?” But the multitude of defini-
tions has proliferated to challenge stereotypical views
about women that have become entrenched in culture.
Spelman (1988, 134) writes, “’Women’ are what fe-
males of the human species become, or are supposed
to become, through learning how to think, act, and
live in certain ways.” In other words, the definition of
woman is disputed because of non-essential cultural
influences imposed upon the female sex.
Because of the fear of essentialism, separating sex
from gender has also been important to feminists, yet
the relationship is nearly inextricable. As commonly
understood, “sex” implies a biological division between
those who have female and male sex organs. “Gender,”
then, reflects the social or cultural influences on sexual
difference. Feminists assert that biological difference is
exploited by patriarchal society to justify fitness for
gender roles, which almost exclusively place women
subordinate to men. Frye (2005, 44) points out that es-
sentialism permeates classification, but “the word fe-
male has been virtually banished from the discourse
of…feminist theory” because of fear of biological de-
terminism. Similarly, Olson (2007) argues that despite
Plato’s intentions, the A/Not A division often morphs
into A/B, with B representing the opposite of A. If A
is “men,” they automatically may be culturally ascribed
conditions such as strength and rationalism, which
means “women” involuntarily become B, with traits of
weakness and emotion. When group members do not
possess, do not believe they possess, or do not want
the traits associated with the group, the dilemma be-
comes, as Alcoff (1995, 434) notes, “our very self-
definition is grounded in a concept that we must de-
construct and de-essentialize in all of its aspects.”
Women do not want to be pigeonholed as possessing
one trait or another, as yet by falling under the name of
“woman,” something must connect them, and most
often that something has been the female sex.
Complicating the sex-gender division are the gen-
der and sex identities that differ from or resist the
heteronormative or mainstream binary biological di-
visions. The biological division between sexes is not
nearly as clear-cut as most people assume. Research
surveying medical literature between 1953 and 2000
reports that “1.7 % of live births do not conform to
the Platonic ideal of absolute sex chromosomal, go-
nadal, genital and hormonal dimorphism.” They also
estimate that about 1-2 out of every 1000 live births
undergo “corrective” surgery in order to conform the
genitals to one sex or the other, which can cause gen-
der confusion later in life (Blackless, et al. 2000, 161).
Because of discomfort with categorical ambiguity,
Hale (1996, 288) writes that babies “born with ‘am-
biguous’ genitals are assigned to a sex as soon as pos-
sible,” usually before eighteen months old. Further-
more, Halberstam (1994) identifies a multitude of
self-categorizations within the queer community that
explode the binaries of homosexual/heterosexual, as
well as male/female and man/woman.
4.0 Applying Prototype Theory to Sex and Gender
Does defining a concept automatically seal the
boundaries of the category? Chow (1993, 15, empha-
sis original) asks, “isn't "naming" precisely the center-
ing, the essentializing act?” Chow’s use of the word
“centering” suggests prototype theory, where a center
exists, but with permeable rather than rigid bounda-
ries. If feminists and queer theorists reject the hierar-
chical model of concepts because it subordinates and
traps them within someone else’s structure, why not
try a different model that rejects essentialism and mu-
tual exclusivity? Although gender and sex are fluid in
reality, current cultural standards are constrained by
the man/woman binary, so those of ambiguous gender
generally end up in one category or another based on
outward appearance. However, Haslanger (2000, 38)
writes that if gender is considered “social position, we
must allow that one can be a woman without ever act-
ing … [or] feeling like a woman, or even having a fe-
male body.” A prototype conceptualization would
find that indeed a network of qualities common to
“woman” exist, yet none of those qualities are re-
quired or even central, provided some of the other fea-
tures typical to the class are present. It should be
noted that possession of the qualities does not auto-
matically place one in the category “woman,” i.e. a
male in an occupation acceptable for a woman does
not make him a woman. A woman usually, but not al-
ways dresses in a feminine fashion. A woman usually,
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
332
but not always, has female sex organs. Thus, the con-
cept avoids essentialist determinism, yet acknowledges
similarity.
4.1 Hale’s “Defining Characteristics of
the Category Woman”
What are the qualities that might work for a concep-
tualization of “woman?” Although not intentionally
working within a prototype framework, Hale (1996,
290) provides 13 characteristics that could determine
membership in the category “woman,” but specifically
states, “None of these characteristics is a necessary or
sufficient condition.” The first five cover characteris-
tics typically associated with sex. Hale (290-91)
weights these qualities more heavily, since culturally,
the presence of biological sex is considered the most
important:
1. Absence of a penis;
2. Presence of breasts;
3. Presence of reproductive organs;
4. Presence of estrogen and progesterone within the
appropriate range for age; and,
5. Presence of XX or absence of Y chromosomes.
The next cluster of characteristics refer to cultural ex-
pectations for the gender of woman and the idea of
identity. Hale asks, “Do you feel yourself to be a
woman? Then, according to this defining characteris-
tic, you are” (292):
1. Having a gender identity as a woman;
2. Having an occupation considered acceptable for a
woman;
3. Engaging in leisure pursuits considered to be ac-
ceptable for a woman; and,
4. Engaging in a form of sexual/affectional relation-
ship with a man commonly recognized as a hetero-
sexual, who does not identify as gay or bisexual.
The next set (293) has to do with looking and acting
like a woman according to cultural norms. This is the
most difficult to achieve for male-to-female trans-
genders who strive to “pass” as a woman:
1. Achieving and maintaining a physical gender self-
presentation of a woman;
2. Behaving in ways that produce the gender assign-
ment “woman” to those with whom one interacts;
and,
3. Giving textual clues to produce the gender assign-
ment “woman,” such as using female pronouns to
refer to oneself, having documents that bear the
designation “F,” and having a feminine name.
Hale asks whether people with whom one interacts
unambiguously think she’s a woman? If so, she’s a
woman. And finally:
Having a history consistent with the gender as-
signment “woman” that provides an unbroken
line from female infancy to womanhood.
Collapsing together the social and the biological,
“woman” becomes more inclusive of identities lo-
cated on the border of gender categories.
5.0 Discussion/Implications
But what if one believes that “woman” cannot possi-
bly have a penis? The inclusiveness afforded by pro-
totypes eliminates the either/or binary constraint
demanded by classical concept theory; however, the
individual nature backfires where belief systems ex-
clude particular traits. Prototypes are formed in any
number of ways—frequency of perception, personal
preferences and beliefs, culture, goals, and experience,
among others. Indeed, an issue with prototype theory
is the notion of the prototype as an ideal representa-
tive of the set. Rosch (1999, 65, 71) places value
judgments on members of the group by grading them
as “good” representatives (closer family resemblance
to the prototype) or “bad” (least resemblance to pro-
totype). This could lead to devaluation of instantia-
tions that do not closely match that prototype, par-
ticularly if the “paragon” approach is used. In the case
of gender, the “average” rather than “exemplar” ver-
sion of a prototype would work best, since no ideal
exemplar exists. Hale (1996, 283), too, warns that
women behaving in ways not compatible with a para-
gon may not be considered “a real woman.”
And what is meant by “acceptable for a woman?”
What is “acceptable for a woman” is tied to current
cultural expectations, rendering the concept contex-
tually dynamic, but circular. Rosch (1999, 71) calls
the formation of concepts “situation based and par-
ticipatory.” Therefore, a more inclusive prototype re-
lies on the provision of models for people to perceive
that influence their individual prototypes. In a ho-
mogenous society that lacks diversity of imagery or
messages, what is “acceptable for a woman” may be
limited, but other situational factors can neutralize
the dominant impressions. In persistently oppressive
societies or in situations where individuals live in iso-
lation from outside influence, conceptualizations
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
333
might resist change, but as a whole, prototype theory
would allow for dynamic, historically current con-
cepts. Contrasting messages would be most evident
and influential in societies where widespread access to
information is available.
6.0 Conclusion
Prototype theory might work on a theoretical or psy-
chological plane, but could it work as a technique for
knowledge organization? The dynamic and personal
nature of prototypes presents some problems for
classification systems that require brevity in concep-
tual representation and some degree of stability. As
normative notions, concepts still can be subject to the
“tyranny of the majority,” or risk being constructed
by those in power, similar to other concept theories.
According to prototype theory, perception truly is re-
ality. This means that prototypical definitions rely
heavily on context and are significantly subject to so-
cial flux. Bowker and Star (1999, 106) believe that
medical classifications are “’naturally’ prototypical”
but they must “appear Aristotelian to bear the bu-
reaucratic burden that is put on them.” The same can
be said about library classification. However proto-
typical the concepts are, the underlying goal is to
place one book in one place.
Can the Western mind comprehend such an inclu-
sive theory that allows such nebulous characteriza-
tions? Does the vagueness of the group boundaries
and moment-by-moment changeability render con-
cepts meaningless? Does it mean that anyone can be-
lieve anything he or she wants? As human constructs,
even in a classical framework, conceptualizations are
already perception based and context sensitive, as so
much classification research has shown. Since proto-
type theory is based on perception, adding an impos-
sibility such as the ability to run 500 miles per hour,
would unlikely be accepted by any others. That being
said, prototypes need not be experienced to be ac-
cepted; otherwise, any new instances could not be ab-
sorbed or learned (such as seeing an ostrich for the
first time and recognizing it as a bird). Minority views
most likely would find difficulty gaining credence,
but as a socially constructed space, if one makes an
association, it can be so. In these cases, self-determi-
nation gains importance. Acts of rebellion and resis-
tance can bring non-mainstream perceptions closer to
a prototype. Different ideological conceptualizations
could be equally valid, which can be both valuable and
dangerous.
The concepts of sex and gender as imagined using
the prototype theory of concepts as a theoretical
framework provides structural flexibility and inclu-
siveness. The “family resemblance” stabilizes terms
enough to recognize the overarching idea, allows va-
riation in particular instantiations and accommodates
overlap. However, does it provide enough stability to
manage the “bureaucratic burden” required of it?
Does the contextual sensitivity destroy any chance at
constancy? Can it complement classical theory with-
out losing the flexibility that makes it valuable? While
prototype theory might work in theory, it can be ma-
nipulated to advance offensive definitions as easily as
desirable definitions. Further investigation is neces-
sary to determine whether those obstacles can be
overcome and how that flexibility could be sustained
in a practical environment.
References
Alcoff, Linda. M. 2005. The metaphysics of gender
and sexual difference. In B.S. Andrew, J. Keller, and
L.H. Schwartzman, eds., Feminist interventions in
ethics and politics: Feminist ethics and social theory.
Lanham, MD: Rowman & Littlefield, pp.17-39.
Blackless, Melanie, Charuvastra, Anthony, Derryck,
Amanda, Fausto-Sterling, Anne, Lauzanne, Karl,
and Ellen Lee. 2000. How sexually dimorphic are
we? Review and synthesis. American journal of hu-
man biology 12: 151-66.
Dahlberg, Ingetraut. 1995. Conceptual structures and
systemization. International forum on information
and documentation 20(3): 9-24.
Frye, Marilyn. 2005. Categories in distress. In Femi-
nist interventions in ethics and politics: feminist eth-
ics and social theory, B.S. Andrew, J. Keller, and
L.H. Schwartzman, eds. Lanham, MA: Rowman &
Littlefield, pp. 41-58.
Hale, Jacob. 1996. Are lesbians women? Hypatia
11(2): 94-121.
Haslanger, Sally. 2000. Gender and race: (What) are
they? (What) do we want them to be? NOÛS
34(1): 31-55.
Hjørland, Birger. 2009. Concept theory. Journal of the
American Society for Information Science and Tech-
nology 60: 1519-36.
Jacob, Elin. K. 2004. Classification and categoriza-
tion: A difference that makes a difference. Library
trends 52(3): 515-40.
Margolis, Eric, and Laurence, Stephen. 2006. Con-
cepts. Stanford encyclopedia of philosophy. http://
plato.stanford.edu/entries/concepts.
Knowl. Org. 38(2011)No.4
M. J. Fox. Prototype Theory: An Alternative Concept Theory for Categorizing Sex and Gender?
334
Nye, Andrea. 1990. Words of power: A feminist read-
ing of the history of logic. New York: Routledge.
Olson, Hope A. 1999. Exclusivity, teleology and hier-
archy: Our Aristotelian legacy. Knowledge organiza-
tion 26: 65-73.
Olson, Hope A. 2001. Sameness and difference: A
cultural foundation of classification. Library re-
sources & technical services 45: 115-22.
Olson, Hope A. 2007. How we construct subjects: A
feminist analysis. Library trends 56(2): 509-41.
Rosch, Eleanor. 1999. Reclaiming cognition: The pri-
macy of action, intention and emotion. The journal
of consciousness studies 6(11-12): 61-77.
Slaughter, Mary M. 1982. Universal languages and sci-
entific taxonomy in the seventeenth century. New
York: Cambridge UP.
Szostak, Rick. 2004. Classifying science: phenomena,
data, theory, method, practice. Norwell, MA:
Springer.
Szostak, Rick. 2010. Comment on Hjørland’s concept
theory. Journal of the American Society for Informa-
tion Science and Technology 61: 1076-77.
Wittgenstein, L. [1988] 1953. Philosophical investiga-
tions, 3
rd ed. Trans. G.E.M Anscombe. London:
Basil Blackwell.