KARENBARAD

MEETING THE UNIVERSE HALFWAY:

REALISM AND SOCIAL CONSTRUCTIVISM WITHOUT
CONTRADICTION

Because truths we don't suspect have a hard time

making themselves felt, as when thirteen species
of whiptaillizards composed entirely of females
stay undiscovered due to bias
against such things existing,
we have to meet the universe halfway.
Nothing will unfold for us unless we move toward what
looks to us like nothing: faith is a cascade.
The sky's high solid is anything
but, the sun going under hasn't
budged, and if death divests the self
it's the sole event in nature
that's exactly what it seems.
[From the poem "Cascade Experiment", by Alice Fulton (Fulton, 1990)]

1. INTRODUCTION

The morning after giving an invited lecture on the socially constructed nature of
scientific knowledge, I had the privilege of watching as a STM (scanning tunneling
microscope) operator zoomed in on a sample of graphite, and as we approached
a scale of thousands of nanometers ... hundreds of nanometers ... tens of nano-
meters ... down to fractions of a nanometer, individual carbon atoms were imaged
before our very eyes. The experience was so sublime that it sent chills through my
body - and I stood there, a theoretical physicist who, like most of my kind, rarely
ventures into the basements of physics buildings experimental colleagues call
"home", conscious that this was one of those life moments when the amorphous
jumble of history seems to crystallize in a single instant. How many times had I
recounted for my students the evidence for the existence of atoms? And there they
were - just the right size and grouped in a hexagonal structure with the interatomic
spacings as predicted by theory! "If only Einstein, Rutherford, Bohr, and especially
Mach, could have seen this!" I found myself exclaiming. And as the undergraduate
students operating the instrument (that they had just gotten to work the day before
by carefully eliminating sources of vibrational interference - we're talking nano-
meters here!), disassembled the chamber which held the sample so that I could see
for myself the delicate positioning of the probe above the graphite surface, expertly
161
L. H. Nelson and J. Nelson (eds), Feminism, Science, and the Philosophy of Science, 161-194.
© 1996 Kluwer Academic Publishers.
162 KARENBARAD

cleaved with a piece of scotch tape, I mused outloud that "seeing" atoms will
quickly become routine for students (as previous generations in turn found the
examination of cells by visual light microscopes to be and then the structure of
molecules by electron microscopes so) and that I was grateful to have been brought
up in a scientific era without this particular expectation.
At this point in my story, I imagine there will be scientific colleagues who will
wonder whether this presented a moment of intellectual embarrassment for your
narrator who had on the previous night insisted on the socially constructed nature
of scientific knowledge. In fact, although I was profoundly moved by the event I
had just witnessed, standing there before the altar of the efficacy of the scientific
enterprise, I was unrepentant. For as social constructivists have tried to make clear,
empirical adequacy is not an argument that can be used to silence charges of con-
structivism. The fact that scientific knowledge is socially constructed does not
imply that science doesn't "work", and the fact that science "works" does not mean
that we have discovered human-independent facts about nature. (Of course, the fact
that empirical adequacy is not proof of realism is not the endpoint, but the starting
point for constructivists, who must explain how it is that our constructions work -
an obligation that seems all the more urgent in the face of increasingly compelling
evidence that the social practice of science is conceptually, methodologically, and
epistemologically allied along particular axes of power.')
On the other hand, I stand in sympathy with my scientific colleagues who want
science studies scholars to remember that there are cultural and natural/material
causes for knowledge claims. While most social constructivists go out of their way
to attempt to dispel the fears that they are either denying the existence of a human-
independent world or the importance of material factors in the construction of
scientific knowledge, the bulk of the attention has been on cultural factors. To be
fair, this is where the burden of proof has been placed: social constructivists have
been responding to the challenge to demonstrate the falsity of the worldview that
takes science as the mirror of nature. Nonetheless, as both the range and sophis-
tication of constructivist arguments have grown, the charge that they embrace an
equally extreme position - that science mirrors culture - has been levied against
them with increasing vigor. While few constructivists actually take such an extreme
position, we would be remiss in simply dismissing this charge as a trivial over-
simplification and misunderstanding of the varied and complex positions that come
under the rubric of constructivism. For the anxiety being expressed, though admit-
tedly displaced, touches upon the legitimate concern about the privileging of epi-
stemological issues over ontological ones in the constructivist literature. Ontological
issues have not been totally ignored, but they have been overshadowed.
The ontology of the world is a matter of discovery for the traditional realist. The
assumed one-to-one correspondence between scientific theories and reality is used
to bolster the further assumption that scientific entities are unmarked by the dis-
coverers: that is, nature is taken to be transparently given. Acknowledging the
importance of Cartwright's (1983) philosophical analysis decoupling these assump-
tions and her subsequent separation of scientific realism into two independent
 MEETING THE UNIVERSE HALFWAY 163
positions - realism about theories and realism about entities - Hacking (1982), like
Cartwright, advocates realism towards entities. Shifting the traditional emphasis in
science studies away from theory construction to the examination of experimental
practice, Hacking grounds his position on the ability of the experimenter to man-
ipulate entities in the laboratory. Galison (1987) also centers experimental practice
in his constructivist analysis comparing three different periods of twentieth-century
physics experimentation, wherein he generalizes Hacking's criterion for the reality
of entities by underlying the importance of the notions of stability (i.e., invariances
of results under changing experimental conditions, rather than the narrower cat-
egory of manipulation) and directness (i.e., epistemologically, but not necessarily
logically, non-inferential). There are other constructivist approaches which go
further in interrogating the transparency of our representations of nature. Latour
(1993) prioritizes stability as well, posing it as one variable of a two-dimensional
geometry whose other axis connects the poles of Nature and Society. Essence then
becomes the trajectory of stabilization within this geometry that is meant to charac-
terize the variable ontologies of quasi-objects. In contrast, Haraway (1988) em-
phasizes instability: it is the instability of boundaries defining objects that is the
focal point of her explicit challenge not only to conceptions of nature that claim to
be outside of culture, but also to the separation of epistemology from ontology.
Interestingly, the instability of boundaries and Haraway's insistence that the objects
of knowledge are agents in the production of knowledge, feature her notions of
cyborgs (1985) and material-semiotic actors (1988) which strike up dissonant and
harmonic resonances with Latour's hybrids and quasi-objects (1993). Moving to
what some consider the opposite pole of the traditional realist position is the post-
structuralist position. To many scientists as well as science studies scholars,
Derridian forms of poststructuralism that disconnect sign from signified seem to be
the ultimate in linguistic narcissism. While insisting that we are always already in
the "theater of representation", Hayles (1993) takes exception to extreme views that
hold that language is groundless play, and while she does not provide us with
access to the real she does attempt to place language in touch with reality by re-
conceptualizing referentiality. Hayles' theory of constrained constructivism (1993)
relies on consistency (in opposition to the realist notion of congruence) and the
semiotic notion of negativity to acknowledge the importance of constraints offered
by a reality that cannot be seen in its positivity: as she puts it, "Although there may
be no outside that we can know, there is a boundary" (p. 40, original emphasis).
These attempts to say something about the ontology of our world are exceptions
rather than the rule in the constructivist literature. There is a need to elaborate
further upon the crafting of ontologies. We need to understand the technologies by
which nature and culture interact. Does nature provide some template that gets
filled in by culture in ways that are compatible with local discourses? Or do specific
discourses provide the lenses through which we view the layering of culture upon
nature? Does the full "texture" of nature get through or is it partially obliterated or
distorted in the process? Is reality an amorphous blob that is structured by human
discourses and interactions? Or does it have some complicated irregular shape that
164 KARENBARAD

is differently sampled by varying frameworks that happen to "fit" in local regions

like coincident segments of interlocking puzzle pieces? Or is the geometry fractal
so that it is impossible for theories to match reality even locally? At what level of
detail can any such question be answered, if at all? And what would it mean? Is it
possible to take any of these questions seriously within the academy, in the U.S., in
the late twentieth century? Won't this still sound too much like metaphysics to
those trained during the various states of decay of positivist culture? And if we
don't ask these questions what will be the consequences? For as Donna Haraway
reminds us, "what counts as an object is precisely what world history turns out
to be about" (1988, 588). I seek some way of trying to understand the nature of the
interplay of the material and the cultural in the crafting of an ontology. Con-
sequently, I will place considerably more emphasis on ontological issues than is
common in science studies, although I will not ignore the epistemological issues
either, since like Haraway's material-semiotic actors, the ontology that I will offer
is not outside of epistemology.
Upon articulating a new ontological and epistemological framework, I will own
up to its realist tenor. After a resurgence of interest in scientific realism in the
1980s, its popularity seems to have waned once again, if not the result of the
deathknell sounded by Fine's (1984) clever according of the metatheoretical failure
of arguments for realism, then at least by the commonplace tendency on the part of
constructivists to present scientific realism as naive, unreflexive, and politically
invested in its pretense to assume an apolitical posture. In fact, the pairing of social
constructivism with some form of antirealism has come to seem almost axiomatic:
if we acknowledge the cultural specificity of scientific knowledge construction, are
we not obligated to relinquish the hope of constructing theories that are true repre-
sentations of independent reality? For example, in offering a concrete case of the
underdetermination thesis, Cushing (1994) argues that the fact that distinctive theo-
ries can account for the same empirical evidence means that realists are hard-pressed
to make an argument for theoretical access to the actual ontology of our world. 2 For
the most part, social constructivists have expressed either outright disdain for or at
least suspicion towards realism, and have explicitly adopted antirealist positions, or
they have refused the realism-antirealism debate altogether either because they feel
limited by this very opposition (see for example Fine, 1984; Pickering, 1994) or they
have thought it more fruitful to focus on other issues. As an admitted social construc-
tivist, I must confess to having sympathy with all of these positions, but I do not want
to deny my own realist tendencies or the realist features of the framework I present.
While I acknowledge that realism has been invoked to support both oppressive and
liberatory positions and projects, my hope is that at this historical juncture, the weight
of realism - the serious business and related responsibility involved in truth hunting -
can offer a possible ballast against the persistent positivist scientific culture that too
easily confuses theory with play (see Barad, forthcoming).
Realizing the multiplicity of meanings that realism connotes, at this juncture I
want to clarify how I take realism in the first instance. As a starting point, I follow
Cushing's lead:
 MEETING THE UNIVERSE HALFWAY 165
I assume, perhaps unreasonably, that a scientific realist believes successful scientific theories to be
capable of providing reliable and understandable access to the ontology of the world. If one weakens
this demand too much, not much remains, except a belief in the existence of an objective reality to
which we have little access and whose representation by our theones is nebulous beyond meaningful
comprehension. In such a situation, is it worth worrying about whether or not one is a realist? (Cushing,
1994,270, fn. 26).

Although 1 will ultimately add substantive qualifications to this definition, 1 do not

intend to weaken what 1 take to be the spirit of this demand, and 1 therefore have
selected this starting point to clarify the sense of realism with which 1 mean to
engage, as separate from some other more general uses in the science studies litera-
ture, including discussions that oppose realism to relativism, or realism to linguistic
monism, or realism to subjectivism. My first concern is not with realism in these
senses: 1 grant that there are forms of antirealism that are not relativist, that do not
deny the existence of an extralinguistic reality, and that are compatible with various
notions of objectivity. That is, in the spirit of Cushing's query, 1 want to limit the
elasticity of the meaning of realism for my initial purposes. Science studies schol-
ars have labored long and hard to articulate moderate social constructivist positions
that reject the extremes of objectivist, subjectivist, absolutist, and relativist stances,
but it is perhaps inappropriate to label these as realist on just such bases alone. That
is, 1 do not want to tum these accomplishments aside by setting up realism as the
foil to the entire family of apparitions, including some that scientists find most
haunting. For example, feminist science studies scholars in particular have over-
whelmingly rejected the specter of epistemological relativism, with an intensity
shared by scientists (a fact which may come as a surprise to scientists who have not
studied the feminist literature). Seeing epistemological relativism as the mirror twin
of objectivism, and both as attempts to deny the embodiment of knowledge claims,
feminist theories of science including Haraway's theory of situated knowledges
(1988), Harding's strong objectivity (1991), Keller's dynamic objectivity (1985),
and Longino's contextual empiricism (1990) articulate nonrelativist constructivist
positions. Consequently, although my discussion of realism in this paper is con-
cerned with the sense in which access to the ontology of our world is possible, ad-
ditionally 1 will also attempt to satisfy the high standards that have already been set
by specifying the ways in which the new form of realism that 1 propose rejects
these other extreme oppositions. 1 use the same label, "agential realism", for both
the new form of realism and the larger epistemological and ontological frame-
work that 1 propose. (My motivation for using an adjectival form of agency as the
modifier will be clarified later.)

2. AGENTIAL REALISM: AN OVERVIEW

The inspiration for agential realism comes from my reading of Niels Bohr's philo-
sophy-physics. (I use this hyphenated structure, instead of the usual "philosophy
of physics", to emphasize Bohr's unwillingness to think of these interests as
distinctive in any sense, contrary to the sharp disciplinary boundaries that are
166 KARENBARAD

important to contemporary physics culture (Barad, 1995).) Bohr's philosophy-

physics provides a fruitful starting point because it involves a critical examination
of observation/measurement processes: in contrast to the inconsequential role of the
observer in Newtonian physics, Bohr argued that quantum physics requires a new
logical framework that takes the observation processes into account. Measurement
is a potent moment in the construction of scientific knowledge - it is an instance
where matter and meaning meet in a very literal sense. For example, in the context
of studies of the practice of experimental high-energy physics, science studies
scholars have emphasized the role of detectors as sites for making meaning
(Traweek, 1988; Galison, 1987; Pickering, 1984). My focus here is on the em-
bodiment of culture within theory. That is, I read Bohr's philosophy-physics as an
argument for the necessity of including practice within theory: that, contrary to tra-
ditional views of physical theory that take the actual practice of measurement to be
outside of theory, and according to the logical positivist/empiricist program which
assumes that measurements transparently adjudicate among theories, Bohr situates
practice within theory, since to ignore practice is to get the theory wrong. This is
not to suggest that all is reduced to theory, but that theory, as a matter of principle,
must itself be embodied in practice and cannot abstract itself from these issues. 3
While I fully suspect postmodern readers to be readily suspicious of theoretical
moves that elevate practical issues to the realm of principles, I will show that this
implicit universality amounts to the common constructivist assertion that all knowl-
edges are local knowledges. That is, I will indicate how this theoretical analysis of
measurement can be understood as the literal embodiment of objectivity in the
sense of Haraway's theory of situated knowledges (1988; see also Barad, 1996).
Now I am quite aware that the ubiquitous appropriation of quantum theory
makes it dangerous material to handle these days, and the addition of feminist
theory to my list of concerns seems to be quite enough to detonate the explosive
mixture, so a few preliminary words of caution are in order. In a sense, to accom-
plish my task I need to "rescue" quantum theory from both its overzealous ad-
vocates and its unreflective practitioners. In the popular literature quantum physics
is often positioned as the scientific path leading out of the West to the metaphysical
garden of Eastern mysticism. Paralleling these popular renditions, one can find sug-
gestions in the feminist literature that quantum physics is inherently less andro-
centric, more feminine, and generally less regressive than the masculinist ten-
dencies found in Newtonian physics. But those who naively embrace quantum
physics as some exotic Other that will save our weary Western souls forget too
quickly that quantum physics underlies the workings of the A-bomb, that particle
physics (which relies on quantum theory) is the ultimate manifestation of the ten-
dency towards scientific reductionism, and that quantum theory in all its applica-
tions continues to be the purview of a small group of primarily Western-trained
males. It is not my intention to contribute to the romanticizing or mysticizing of
quantum theory. On the contrary, as a physicist, I am interested in engaging in a
rigorous dialogue about particular aspects of specific discourses on quantum
physics and the implications. Similarly, I do not make any claims here about Niels
 MEETING THE UNIVERSE HALFWAY 167
Bohr being an unappreciated or closet feminist. Nor is my aim to critique physics by
holding it up to some fixed notion of gender. On the contrary, the feminist analysis I
present here simultaneously interrogates the notions of identity and science. 4
On the other hand, I part company with my neo-positivist physics colleagues
who believe that philosophical concerns are superfluous to the real subject matter of
physics. Indeed, I am sympathetic to Bohr's view that philosophy is integral to
physics. Niels Bohr was one of the most important physicists of the twentieth
century, and his "philosophical" writings span a period of approximately four
decades. 5 Bohr is considered to be the primary author of the so-called Copenhagen
interpretation of quantum mechanics. 6 Although alternative interpretations
have been advanced ever since the formulation of quantum theory during the mid-
1920s, the physics community claims allegiance to the Copenhagen interpretation.?
Unfortunately, the vast majority of physicists have no more than a passing interest
in the philosophical issues, and prefer to focus on the powerful tools that the
quantum formulation provides for purposes of calculation. This avoidance has had
its cost: the foundational issues of this fundamental physical theory remain
unresolved and the culture of physics is such that unreflective attitudes and
approaches are rewarded. While I will not make any arguments about the superior-
ity of Bohr's approach to quantum physics, the simultaneous centrality and
marginality of his approach makes it particularly interesting. 8
Bohr often makes reference to the epistemological lessons of quantum theory,
and he sees the framework that he offers for quantum physics as having general
relevance beyond physics (Folse, 1985). So it is not at all inappropriate that at-
tention has been given to the larger philosophical implications of Bohr's philosophy-
physics, leaving specific issues surrounding the interpretation of quantum theory
aside. My approach will be to draw out the specifics of a consistent Bohrian
framework, grounding the analysis in the physics, and then to consider the larger
implications.
The first task is necessary since there is much disagreement in the secondary
literature about how to interpret Bohr. For example, Bohr has been called a
positivist, an idealist, an instrumentalist, a (macro)phenomenalist, an operationalist,
a pragmatist, a (neo )Kantian , and a realist by various authors. One of the difficulties
in assigning a traditional label to Bohr's interpretative framework is the fact that
Bohr is not specific about his ontological commitments. In order to fill this crucial
gap, I propose an ontology that I believe to be consistent with Bohr's views,
although I make no claim that this is what he necessarily had in mind. That is, as a
result of Bohr's inattentiveness to concerns of ontology, there will be vast dif-
ferences in opinion about this matter, and I am less interested in trying to figure out
what Bohr was actually thinking than what makes sense in the context of what
Bohr does tell us. My approach is to use Bohr's writings as the context for my
thinking about these issues; I do not take them as scripture (see Methodological
Interlude). Using this analysis of Bohr's philosophy-physics as inspiration, I
introduce agential realism as a framework that ties together the epistemological and
ontological issues.
168 KARENBARAD

I then show how agential realism can be used to address particular concerns that
social constructivist approaches to science make apparent, including some of the
ones enumerated in the previous section. I diverge from Bohr in strategy here, but
not in spirit. Bohr's methodological approach was to draw out the epistemological
lessons of quantum theory for other fields of knowledge by essentially trying to
guess what the relevant Complementary variables would be in each arena. This
analogic strategy often failed: both because he proposed a set of variables that
turned out not to be Complementary, and because the implications drawn on this
basis watered down the complexity and richness of the "epistemologicallessons".9
My approach will be to examine specific implications by directly taking on a dif-
ferent set of epistemological and ontological commitments. That is, I will not use
the notion of Complementarity as a springboard; instead I directly interrogate
particular philosophical background assumptions that underlie specific concerns.
Finally, I want to make explicit the distinction between my approach and a host
of analogical (mis)appropriations of quantum theory that are more common in the
literature than physicists would wish. I will not put forward any argument to the
effect that quantum theory of the microworld is analogous to situations that interest
us in the macroworld - be they religious, spiritual, psychological, or even those en-
countered in science studies. My focus is on the development of widely applicable
epistemological and ontological issues, that can be usefully investigated by a rigor-
ous examination of measurement processes as explicated by Bohr's understanding
of quantum physics. To ask whether it is not suspect to apply arguments made
specifically for microscopic entities to the macroscopic world is, in this case, to
mistake the approach as analogical. The epistemological and ontological issues are
not circumscribed by the size of Planck's constant (see note 12). That is, I am not
interested in mere analogies but rather widely applicable philosophical issues such
as the conditions for objectivity, the appropriate referent for empirical attributes,
the role of natural as well as cultural factors in scientific knowledge production, and
the efficacy of science.

3. MEASUREMENT MATTERS

Often the development of physics has taught us that a consistent application of even the most elementary
concepts indispensable for the description of daily experience, is based on assumptions initially un-
noticed, the explicit consideration of which is, however, essential if we wish to obtain a classificatIOn of
more extended domains of experience as clear and as free from arbitranness as possible .... This de-
velopment has contributed to the general philosophical clarification of the principles underlying human
knowledge (Bohr. 1937.289-290).

Classical epistemological and ontological assumptions, such as the ones found to

underlie Newtonian physics, include an autonomously existing world that is de-
scribable independently of our experimental investigations of it. This accounts for
the fact that the process of measurement is transparent and external to the discourse
of Newtonian science. It is assumed that objects and observers occupy physically
and conceptually separable positions. Objects are assumed to possess well-defined
intrinsic attributes, and it is the job of the scientist to cleverly discern these inherent
 MEETING THE UNIVERSE HALFWAY 169
characteristics, obtaining the values of the corresponding context-independent
variables through some benignly invasive measurement procedure. The repro-
ducibility of measured values under the methodology of controlled experimentation
is used in support of the objectivist claim that what has been obtained is an objec-
tive representation of intrinsic properties that characterize the objects of an un-
controlled, independent reality. IO The transparency of the measurement process
in Newtonian physics is a root cause of its value to and prestige within the Enlight-
enment culture of objectivism.
The two basic assumptions of measurement transparency underlying Newtonian
physics that were challenged by Bohr's interpretation of quantum physics are: II
(1) Measurements involve continuous, determinable interactions; that is, an un-
ambiguous, inherent, Cartesian-like cut between knower and known delineates
object from observational apparatus.
(2) The applicability of conceptual schema is independent of measurement
processes; concepts are abstractable, universal, definite, and context-
independent.
The hallmark of Newtonian physics is its determinism, its proclaimed ability to
predict and retrodict the full set of physical states of a system for all time: given the
initial conditions (i.e., the simultaneous specification of position and momentum at
one instant in time), entire particle trajectories can be calculated. In the Newtonian
picture, the position and momentum of an object can be determined by a time-of-
flight measurement, for example, in which light impinges on the object and the
scattered light is collected at a detector. Although light has momentum and energy,
the process of illuminating the object can either be made to impart negligible dis-
turbance on the object (intuitively, by continuously lowering the intensity) or the
disturbance can be determined and subtracted out, thereby yielding the desired
values of the position and the momentum of the object as they would have been had
the measurement not been performed. According to Niels Bohr, this objectivist
account of the measurement process rests on false assumptions. 12
Quantum physics is based on an empirically verified discreteness or dis-
continuity (the quantum of action = Planck's constant = h "# 0) in measurement
interactions, initially observed in experiments probing atomic phenomena at the
tum of the century. The lack of continuity means that measurement interactions
cannot be reduced to the point of being negligible and, therefore, determination of
the properties of an independent object are contingent upon subtraction of the
effects of the measurement interaction. Bohr argued that subtraction is impossible;
that the measurement interaction cannot be precisely specified without intervening
in such a way as to disrupt the purpose of the intended measurement. Furthermore,
he argued that the indeterminable discontinuity undermines the separability of the
"object" and the "agencies of observation ".
Bohr's argument for the indeterrninability of measurement interactions is based
on his insistence that concepts are defined by the circumstances required for their
measurement, and therefore mutually exclusive experimental arrangements would
need to be employed simultaneously (which is impossible) in order to determine all
170 KARENBARAD

the features of the measurement interaction. For example, in a time-of-flight meas-

urement used to determine the initial conditions, the momentum imparted by the
light impinging on the object would need to be subtracted out. But a measurement
of the momentum requires an apparatus with movable parts (i.e., the concept
'momentum' is necessarily defined by reference to an apparatus with movable
parts 13), which would then exclude the equally necessary measurement of the po-
sition since position measurements require an apparatus with fixed parts (i.e., the
concept 'position' is necessarily defined by reference to a fixed apparatus).
Therefore, observation is only possible on the condition that the interaction is
indeterminable (i.e., it cannot be subtracted out). Consequently, since observations
involve an indeterminable discontinuous interaction, as a matter of principle, there
is no unambiguous way to differentiate between the "object" and the "agencies of
observation" - no inherent/naturally occurringljixed/universallCartesian cut exists.
Hence, observations do not refer to objects of an independent reality. Bohr's inter-
pretation of quantum theory provides profound challenges to both of the assump-
tions of measurement transparency underlying the Newtonian framework. In fact,
Bohr's philosophy-physics undermines a host of Enlightenment notions, requiring
him to construct a new logical framework (see especially, Folse, 1985), including a
new epistemology, for understanding science.
Bohr moves away from reference to the classical notion of 'disturbance' in
his later writings and emphasizes "quantum wholeness", or the lack of an
inherent/Cartesian distinction between the "object" and the "agencies of observa-
tion", as the central feature of his new descriptive framework. For Bohr, "object"
and "agencies of observation" form a nondualistic whole in the sense that it is con-
ceptually incoherent to refer to an inherent distinction between the two.
"Descriptively, there is a single situation, no part of which can be abstracted out
without running into conflict with other such descriptions (namely, those of com-
plementary situation). The object cannot be ascribed an 'independent reality in the
ordinary physical sense'" (original italics; Hooker, 1972, 156). This is a central
notion in Bohr's philosophy-physics and he uses the term "phenomenon" (in his
later writings) to designate particular instances of wholeness: "While, within the
scope of classical physics, the interaction between object and apparatus can be ne-
glected or, if necessary, compensated for, in quantum physics this interaction thus
forms an inseparable part of the phenomenon. Accordingly, the unambiguous
account of proper quantum phenomena must, in principle, include a description of
all relevant features of the experimental arrangement" (my italics, Bohr, 1963c, 4).
Furthermore, "[t]he essential wholeness of a proper quantum phenomenon finds
logical expression in the circumstance that any attempt at its well-defined sub-
division would require a change in the experimental arrangement incompatible with
the appearance of the phenomenon itself' (Bohr 1963b, 72).
If a cut delineating the "object" from the "agencies of observation" is not
inherent, what sense, if any, should we attribute to the notion of observation? Bohr
suggests that "by an experiment we simply understand an event about which we are
able in an unambiguous way to state the conditions necessary for the reproduction
 MEETING THE UNIVERSE HALFWAY 171
of the phenomena" (quoted in Folse, 1985, 124). The specification of these con-
ditions is tantamount to the introduction of a constructedJagentially positionedJ
movablellocall"Bohrian" distinction between an "object" and the "agencies of
observation ".14 That is, although no inherent distinction exists, every measurement
involves a particular choice of apparatus, providing the conditions necessary to give
definition to a particular set of classical variables, at the exclusion of other essential
variables, and thereby placing a particular constructed cut delineating the "object"
from the "agencies of observation". This particular constructed cut resolves the
ambiguities only for a given context; it marks off and is part of a particular
instance of wholeness, that is, a particular phenomenon.
For example, consider once again an experiment in which light is scattered from
a particle. The scattered light may be directed towards a photographic plate rigidly
fixed in the laboratory and therefore used to record the position, or the light may be
directed towards a piece of equipment with movable parts used to record the
momentum of the scattered light. The first case essentially describes the process of
taking a picture of a particle with a flash camera. In that case, the light is part of the
measuring apparatus. In the latter case, the light's momentum is being measured
and hence it is part of the object in question. ls (This example nicely illustrates the
Bohrian assertion that observation is possible only on the condition that the meas-
urement interaction is indeterminable: since at least one particle of light, or photon,
is required to make a mark on the film recording the position (illustrating the
"quantum discontinuity"), and since the effect that this photon has on the particle
cannot be accounted for unless the photon's momentum is simultaneously deter-
mined, and given that both variables ("position" is definable in the context of an
apparatus with a fixed photographic plate, and "momentum" is definable in the
context of photographic plate on a movable platform) cannot be unambiguously
defined using one particular choice of measuring apparatus, the observation entails
an indeterminable interaction.) Therefore, the measurement of unambiguously
defined quantities is possible through the introduction of a constructed cut which
serves to define "object" and "agencies of observation" in a particular context. 16
Especially in his later writings, Bohr insisted that quantum mechanical measure-
ments are "objective",I7 Since Bohr also emphasized the essential wholeness of
phenomena, he cannot possibly have meant that measurements reveal objective
properties of independent objects. Rather Bohr's use of the term "objectivity" is
tied to the fact that "[n]o explicit reference is made to any individual observer"
(Bohr, quoted in Murdoch, 1987, 99). "Objective" means reproducible and un-
ambiguously communicable - in the sense that "permanent marks ... [are] left on
bodies which define the experimental conditions":

Common to the schools of so-called empirical and critical philosophy. an attitude therefore prevailed of
a more or less vague distinction between objective knowledge and subjective belief. By the lesson re-
garding our position as observers of nature, which the development of physical science in the present
century has given us, a new background has, however, been created just for the use of such words as ob-
jectivity and subjectivity. From a logical standpoint, we can by an objective description only understand
172 KARENBARAD

a communication of experience which does not admit of ambiguity as regards the perception of such
communicatIOns (Bohr, quoted in Folse, 1985, 15).

Clearly, Bohr's notion of "objectivity", which is not predicated on an inherent!

Cartesian distinction between "objects" and "agencies of observation", stands
in stark contrast to any Newtonian sense of "objectivity" denoting observer
independence.
Bohr's term "agencies of observation" is evocative of the central role of agency in
the new epistemological and ontological framework that I will introduce later in this
paper. "Agencies of observation", instead of the more common term "observer",
already signals the inseparability of the material and semiotic apparatuses. That is, in
my reading, a pivotal point in Bohr's analysis is that the physical apparatus (existing
in the realm of classical, macroscopic, everyday, direct experience) marks the con-
ceptual subject-object distinction: the material and semiotic apparatuses form a non-
dualistic whole. In other words, classical descriptive concepts obtain their meaning
by reference to a particular physical apparatus which in tum marks the placement of a
constructed cut between the "object" and the "agencies of observation". Finally, the
point of reference for unambiguous communication is "from permanent marks - such
as a spot on a photographic plate, caused by the impact of an electron - left on the
bodies which define the experimental conditions" (Bohr, 1963c, 3). Therefore,
"bodies which define the experimental conditions" serve as both the endpoint and the
starting point for meaningful observation. For Bohr, measurement and description
entail one another (though not in a narrowly operationalist sense but in the sense of
epistemological indistinguishability).
Quite atypical of the writings of theoretical physicists, Bohr's writings often
include very detailed drawings of experimental apparatuses. As Honner points out
"Bohr insisted on providing elaborate drawings of mechanical devices used for
observing quantum events [in many of his discussions of complementarity], as if to
emphasize the connection between descriptive concepts and classical apparatus"
(Honner, 1987, 119). Though Bohr was a theoretical physicist, he was obsessed
with the details of measurement and was not satisfied to deal with abstract concepts
- for Bohr meaning is tied to the experiential world. 18 (There is historical evidence
that Rutherford, whom some regard as the greatest experimental physicist of this
century, had a profound influence on Bohr, who was a postdoc under Rutherford.)
The question remains: what is the referent for a given objective property (as
unambiguously defined in reference to a given constructed cut) that is obtained by a
given measurement process? Since there is no inherent distinction between object
and instrument, the property which is determined cannot be meaningfully attributed
to either an abstract object or an abstract measuring instrument. That is, the meas-
ured value is neither attributable to an observation-independent object, nor is it a
property created by the act of measurement (which would belie any sensible
meaning of the word "measuremenC).19 My reading is that the measured properties
refer to phenomena, remembering that the crucial identifying feature of phenomena
is that they are particular instances of wholeness, that "the unambiguous account of
proper quantum phenomena must, in principle, include a description of all relevant
features of the experimental arrangement" (Bohr, 1963c, 4).
 MEETING THE UNIVERSE HALFWAY 173
Implicit in our classical descriptive concepts is a subject-object distinction, and
since phenomena entail the placement of a constructed subject-object distinction, it
is consistent to use classical concepts to describe phenomena. In fact, Bohr
strengthens the claim for the appropriateness of our use of classical concepts to
describe phenomena to one of necessity. The following detailed mapping of the
relationship between classical concepts and phenomena can be given to provide
justification for this move: since by their very definition classical descriptive con-
cepts entail a particular subject-object distinction, as specified by the circumstances
required for their measurement, and since phenomena include a constructed
subject-object distinction, namely the one in question that gives definition to a par-
ticular classical concept, it follows that these particular classical concepts are just
the ones that are useful in describing phenomena. 2o That is, phenomena are neces-
sarily described using concepts conditioned by particular subject-object dis-
tinctions. Another way to appreciate the necessity of this condition is that
unambiguous communication necessarily refers to "permanent marks ... left on
bodies", that is macroscopic bodies, that in a particular context are defined as the
"apparatus", and since the "apparatus" in tum specifies the circumstances required
for the definition of particular classical concepts (derived from everyday experience
in the macroscopic world and therefore premised on an object-subject distinction),
it follows that phenomena, which include the particular constructed cut in question,
are necessarily described using classical concepts appropriate to the given con-
text. Again, reference must be made to bodies in order for concepts to have
meaning.
"While in the [classical] mechanical conception of nature, the subject-object dis-
tinction was fixed, room is provided for a wider description through the recognition
that the consequent use of our concepts requires different placings of such a
separation" (Bohr, 1963b, 92). In fact, according to Bohr's Principle of Com-
plementarity all possible ways of drawing the subject-object distinction must be
considered to obtain the maximal accounting of our investigations. That is,
mutually exclusive constructed cuts constituting mutually exclusive experimental
circumstances, thereby agentially manifesting mutually exclusive phenomena serve
to denaturalize the nature of the observational process.
Bohr's epistemological and descriptive framework is radically different from that
associated with Newtonian physics. For Bohr, measurement, far from being
external to the discourse of scientific theories, must play a prominent role in
scientific theorizing: that is, Bohr situates practice within theory. As a result,
method, measurement, description, interpretation, epistemology, and ontology are
not separable considerations. These connections are explored in the sections
following the Methodological Interlude.

4. METHODOLOGICAL INTERLUDE

Einstein once remarked of Bohr, "He utters his opinions like one perpetually groping and never like one
who believes to be in possession of definite truth"' (Einstein, quoted in Pais, 1982, 417).
174 KARENBARAD

Many of the philosophers, historians, and the few physicists who have tried to read
Bohr's works have commented on the difficulty of this task. Bohr's style is atypical
of most science writing. His writing reflects a self-conscious regard of his own de-
scriptive process, which is consistent with his thorough-going examination of the
role of description in scientific knowledge production, fundamental to his approach
to understanding quantum physics. In like manner, I have tried to remain attentive
to my own descriptivelinterpretative process in my reading of Bohr. Consequently,
I make no claims here to have discovered what Bohr was actually thinking or
intending, as separate from my own interpretative apparatus; rather I attempt to
provide a consistent reading within the context of particular ways of resolving
ambiguities. (Recall that for Bohr descriptions refer to phenomena, not to some
independent reality.) There are clear parallels between this methodology and
feminist and other located-know ledges methodologies. This is not mere co-
incidence but, as will become clear later, a reflection of a common critical
reflexivity.
My presentation of the major features of Bohr's post-Newtonian framework and
corresponding epistemology come from more than a decade of extensive study of
Bohr's writings. Interpretative questions about quantum theory plagued me as a
graduate student in theoretical particle physics. (It may seem peculiar to non-
scientists to discover that physics graduate school is not the appropriate context for
engaging such questions. 21 ) By the time I was an assistant professor of physics, my
focus broadened to include the larger philosophical issues in Bohr's post-
Newtonian framework.
The ideas as I have presented them so far are in considerable agreement with
individual features of many of the standard secondary texts on Bohr's philosophy
of physics, including the work of Feyerabend (1962), Hooker (1972), Bohm (1985),
Folse (1985), Petersen (1985), Honner (1987), and Murdoch (1987). It is important
to point out that the views of these scholars are widely divergent on many crucial
points. I do not agree in toto with the views presented in any of these other
accounts, though as I read through the primary texts time and again from the per-
spective of a theoretical particle physicist, various aspects of these works have been
and continue to be helpful to me while I formulate my own evolving views on
Bohr's philosophy-physics.
As a measure of the disagreement among Bohr scholars, consider the question of
the nature of Bohr's interpretative framework. Most Bohr scholars, and many other
scholars who have not studied Bohr, attribute some form of antirealism to Bohr, who
has been called a positivist, an idealist, an instrumentalist, a (macro)phenomenalist,
a relativist, a pragmatist, and a (neo)Kantian. Folse has been one of the strongest
proponents of the minority view that sees Bohr as a realist. As I indicated in the
Introduction, one of the difficulties in resolving the ambiguities in Bohr's position is
that Bohr focuses on epistemological issues in his writings and he never spells out
his ontological commitments. Consequently, it is difficult to discern the nature of
any correspondence he may hold between theory and reality. Without a clear-cut
presentation of a coherent Bohrian ontology, the task of determining what kind of
realist or antirealist position is consistent with Bohr's philosophy-physics seems
 MEETING THE UNIVERSE HALFWAY 175
doomed. In the next section, I present an ontology I believe to be consistent with
Bohr's views, and I then address the question of a correlative interpretative stance.
I will argue that Bohr's philosophy-physics can be understood to be consistent
with a particular form of realism, that I label "agential realism". But at I noted from
the outset, my aim is not so much to provide a faithful representation of Bohr's'
philosophy-physics, as to propose a framework for thinking about critical epistemo-
logical and ontological issues, particularly in science studies. In addressing these
issues in the remainder of the paper it would be just as dishonest to attribute the full
development of this framework to Bohr as it would be to deny that my thinking
about Bohr's philosophy-physics is everywhere present in my formulation.

5. AGENTIAL REALITY AND AGENTIAL REALISM

Bohr has often been badly misunderstood, I believe, because his readers have insisted on reading the
classical ontological and epistemological assumptions into ... [his] remarks ... it presupposes some auto-
nomously existing atomic world which is describable independently of our experimental investigation of
it. There is no such world for Bohr. ... There is no godlike approach possible to the physical world
whereby we may know it as it is "absolutely in itself; rather we are able to know only as much of it as
can be captured in those situations which we can handle conceptually - that is, those situations where
unambiguous commumcation of the results IS possIble .... This is in complete contrast to the classical
realist metaphysics and epistemology where the world is concerned as being the way classical theory
says it is. Independently of our experimental exploratIOn of it. ... (Clifford A. Hooker, 1972, 155-6)

The realism-antirealism distinction is often drawn on the basis of questions about

belief in a correspondence theory of truth, which is rooted in a subject-object I
culture-nature I word-world dualisms. The separation of epistemology from ontology
is a reverberation of these dualism. Bohr's philosophy clearly contests a Cartesian
(inherent, fixed, universal) subject-object distinction, and I will argue here that this
undermines conceptions which see reality as either prior to or outside of language.
What is being described is our participation within nature.
Aage Petersen, in an article entitled "The Philosophy of Niels Bohr", writes:
Traditional philosophy has accustomed us to regard language as something secondary, and reality as
something primary. Bohr considered this attitude toward the relation between language and reality
inappropriate. When one said to him that it cannot be language which is fundamental, but that it must be
reality which. so to speak, lies beneath language. and of which language is a picture, he would reply
"We are suspended In language In such a way that we cannot say what is up and what is down. The word
'reality' is also a word, a word which we must learn to use correctly"22 (Petersen, 1985,302).

In my effort to provide a consistent Bohrian meaning to the term 'reality', I tum to

a very important passage from Bohr's writings: a passage from his response to the
1935 paper of Einstein, Podolsky, and Rosen, the so-called "EPR paper", wherein
Bohr directly challenges the EPR definition of "physical reality".23 Many scholars
have pointed out that the argument Bohr articulates in this passage is pivotal to his
attempt to discredit the analysis of EPR and to resolve the "EPR paradox" once and
for all. I say this both to highlight the fact that I have not chosen some obscure or
arbitrary passage from Bohr's writings, but the one in which Bohr has the most at
stake in being careful with the presentation of his ideas on the notion of reality, and
also to express my surprise that none of the scholarship that I have read on Bohr
176 KARENBARAD

emphasizes the positive feature of this passage - that Bohr offers his own definition
of physical reality in the final sentence:
From our point of view we now see that the wording of the above-mentioned criterion of physical reality
proposed by Einstein, Podolsky, and Rosen contains an ambIguity as regards the meaning of the
expression "without in any way disturbing the system". Of course there is in a case like that just con-
sidered no question of a mechanical disturbance of the system under investigation during the last critical
stage of the measuring procedure. But even at this stage there is essentially the question of an influence
on the very conditions which define the possible types of predictions regardinR the future behaviour of
the system. Since these conditions constitute an inherent element of the description of any phenomenon
to which the term "physical reality" can be properly attached, we see that the argumentation of the
mentioned authors does not justify their conclusion that quantum-mechanical description is essentially
incomplete (original italics, Bohr, 1935, 700).

In discussing Bohr's use of the word 'phenomenon' earlier, I pointed out that the
conditions which define the possible types of predictions constitute an inherent
element of the description of any phenomenon. Therefore, the first phrase of the
last sentence is consistent with Bohr's use of the term phenomenon. 24 The last sen-
tence then indicated that the term 'physical reality' can properly be attached to the
phenomenon. Phenomena are constitutive of reality. Reality is not composed of
things-in-themselves or things-behind-phenomena, but things-in-phenomena.
This interpretation is consistent with the following point made by von Weiz-
sacker:
The fact that classical physics breaks down on the quantum level means that we cannot describe atoms
as "little things". This does not seem to be very far from Mach's view that we should not invent "things"
behind the phenomena. But Bohr differs from Mach in maintaining that "phenomena" are always "phe-
nomena involving things", because otherwise the phenomena would not admit of the objectification
without whIch there can be no science of them. For Bohr, the true role of things is that they are not
"behind" but "in phenomena" (quoted in Honner, 1987, IS).

Or as Honner puts it:

The term [phenomenon] was not intended to signify the un interpreted appearance of the object of
experience itself. Nor was Bohr trying to follow the Kantian distinction between the thing-in-itself and
our perception of it. If one wanted to talk about such "things". then they were as Weizsiicker put it, to be
found in the phenomena rather than behind it (Honner, 1987, 68).

The nature of this relationship is a point of contention among Bohr scholars. My

own studies of Bohr's writings brought me to a conclusion similar to von Weiz-
sacker's before I ever started reading any of the secondary texts, and in spite of
subsequent readings of the many different interpretations offered, it has always
seemed very clear to me that this is the only interpretation that respects the complex
intention of the Bohrian notion of 'phenomena'. 25 The point is that phenomena con-
stitute a non-dualistic whole so that it literally makes no sense to talk about inde-
pendently existing things as somehow behind or as the causes of phenomena. A
Bohrian ontology does not entail some fixed notion of being that is prior to
signification (as the classical realist assumes), but neither is being completely
inaccessible to language (as in transcendental idealism) nor completely of language
(as in linguistic monism) - what is being described is our participation within
nature, what I term "agential reality".
 MEETING THE UNIVERSE HALFWAY 177
Bohr often refers to the fact that we are in nature: "In our own century the
immense progress of the sciences has ... given us an unsuspected lesson about our
position as observers o/that nature o/which we are part ourselves" (my emphasis,
Bohr, 1963c, 8). The introduction to the collection Essays 1933-1957 on Atomic
Physics and Human Knowledge begins:
The importance of physical science for the development of general philosophical thinking rests not only
on its contributions to our steadily increasing knowledge of that nature of which we ourselves are part,
but also on the opportunities which time and again it has offered for examination and refinement of our
conceptual tools (my emphasis, Bohr, 1963b, I),

The passage from Bohr's response to EPR continues:

On the contrary, this description, as appears from the preceding discussion, may be characterized as a
rational utilization of all possibilities of unambiguous interpretation of measurements, compatible with
the finIte and uncontrollable mteraction between objects and the measuring instruments in the field of
quantum theory. In fact, it is only the mutual exclusion of any two experimental procedures, permitting
the unambiguous defimtion of complementary physical quantities, which provides room for new phy-
sical laws, the co-existence of which might at first sight appear irreconcilable with the basic principles of
science. It is just this entirely new situation as regards the description of physical phenomena that the
notion of complementarity aims at characterizing (original italics).

Notice that in this last sentence we are told that scientific theories describe physical
phenomena. Since phenomena constitute agential reality, and it is phenomena that
scientific theories describe, it follows that scientific theories describe agential
reality. Were it not for the crucial adjective "agential", emphasizing the non-
objectivist nature of Bohrian ontology, as I've described it here, the conclusion of
this syllogism would sound like the proclamation of a die-hard realist who is ad-
vocating a classical correspondence theory of truth. However, the correspondence
in question is between theories and agential reality, not an observer-independent
reality. Hence, I conclude that Bohr's framework is consistent with a particular
notion of realism, which I label "agential realism". Agential realism is compatible
with the point I made earlier in this section that any notion of realism that is con-
sistent with Bohr's philosophy must not be parasitic on subject-object / culture-
nature / word-world distinctions. 26
That Bohr subscribed to some sort of realism is also supported by his practice of
science. A particularly poignant example of how different philosophical positions
guided the efforts of different segments of the physics community during the 1920s
is given by considering a range of reactions to the notion of "wave/particle duality".
These reactions constitute the twentieth-century contribution to a long historical
debate about the nature of light.
To say that light consists of particles is to insist that light consists of localized
object that occupy a given location at each moment in time. On the other hand, to
say that light consists of waves is to insist that light consists of objects with ex-
tension in space, occupying more than one position at any moment of time, like
ocean waves that move along a stretch of beach; and furthermore, different waves
can overlap and occupy the same position at any moment of time, unlike particles.
Obviously, the concepts of "wave" and "particle" are mutually exclusive: an object
178 KARENBARAD

is either localized or it is extended, it can't be both. And yet, early twentieth-

century experiments seemed to indicate that light behaves as a wave under certain
experimental conditions, and as a particle under a mutually exclusive set of experi-
mental conditions. This result was surprising since in the latter part of the nine-
teenth century the wave theory of light was well confirmed by both theoretical
(Maxwell's electromagnetic theory) and experimental (diffraction and interference
effects) considerations. Hence, a community-wide struggle ensued to resolve this
paradox.
Classical realists hoped to resolve the paradox by finding some unifying ex-
planation. Could it be that all objects are ultimately waves, but that on certain
scales they look like particles? Another type of response came from the
positivists/instrumentals who, like Heisenberg, put their faith in the mathematical
formalism itself and saw the efforts to assign appropriate visualizable concepts to
the mathematics as specious. While this seemed to be a neat and pragmatic solution
to some physicists, others were not so willing to give up on interpretation and
meaning.
Bohr's affinity for some kind of realistic interpretative stance led him to continue
to seek out a solution to this paradox. Bohr participated with a tenacious passion in
the debate. If Bohr had adopted an antirealist attitude it is doubtful that he would
have found it necessary to develop an entirely new approach for understanding
the role of descriptive concepts in science, which became the basis for Com-
plementarity, and ultimately the so-called Copenhagen interpretation of quantum
mechanics. Clearly, interpretative stances matter in the construction of scientific
theories.
A few more historical details may be illuminating here. In 1924, Bohr wrote a
paper with Kramers and Slater that put forth the radical conjecture that perhaps the
most sacred principle in all of physics - the conservation of energy - would have to
be sacrificed at the atomic level in order to find a satisfying resolution of the
wave/particle duality paradox. Surely an instrumentalist or a die-hard antirealist
would not have gone this far in attempting to explain the applicability of dual repre-
sentations. The trio quickly retracted this proposal as soon as contrary empirical
evidence came to light, but Slater never forgave Bohr for convincing him to go
along with such a radical proposal. Bohr then adopted a new approach that entailed
the examination of the circumstances under which these characteristics are manifest
(they only appear under mutually exclusive circumstances), and consequently to an
examination of the context-dependence of descriptive conceptsY Complement-
arity's development was contingent on certain realist commitments on Bohr's
part. Otherwise, Bohr would have been content with the use of alternative de-
scriptions (wave and particle) as evidenced by Heisenberg's instrumentalist
stance.
Furthermore, there is important historical evidence that shows that Bohr strongly
disagreed with Heisenberg about the importance and interpretation of wave/particle
duality. Bohr and Heisenberg went off on separate vacations and developed the
framework of Complementarity and the Uncertainty Principle, respectively; upon
 MEETING THE UNIVERSE HALFWAY 179
returning to Copenhagen, Bohr passionately criticized Heisenberg's derivation of
the Uncertainty Principle for its gross neglect of the centralness of wave/particle
duality for an appropriate analysis. 28
Bohr's interpretative framework deviates in a unique and nontrivial fashion from
classical correspondence or mirroring theories of science. For Bohr, the paradox is
resolved as follows: "wave" and "particle" are classical descriptions that refer to
different mutually exclusive phenomena, and not to independent physical objects.
He emphasized that this saved the theory from inconsistencies since it was impos-
sible to observe particles and wave behaviors simultaneously since mutually
exclusive experimental arrangements are required.
Ambiguity and paradox do not find a Newtonian/Cartesian resolution in this
post-Newtonian framework. No final unifying reductionistic explanation is offered;
only contextual understanding, located knowledges are obtained from the multiple
contestations of the assumption of an inherentifixediuniversallCartesian subject-
object distinction. The ambiguity is only temporarily, contextually decided, and
therefore, descriptive characterizations do not signify properties of abstract objects
or observation-independent beings, but rather describe the "between of our intra-
actions" as it is marked by particular constructed delineations. (Since there is no
sense of two things to interact, I have introduced the term "intra-action" to avoid
reinscription of the contested dichotomy.) In other words, measurements of the
values of the well-defined variables are attributable to the phenomenon as a par-
ticular instance of wholeness, the fully contextual be-in' where the matter and
meaning meet.

6. AGENTIAL REALISM: THE FRAMEWORK

Throughout the field of meanings constituting science, one of the commonalities concerns the status of
any object of knowledge and of related claims about the faithfulness of our accounts to a "real world",
no matter how mediated for us and no matter how complex and contradictory these worlds may be
(Haraway, 1991, 197)

In addition to the question of interpretative stances in science studies, agential

realism provides a framework for addressing broad epistemological and ontological
issues. In this section I develop a few key points that are relevant to the issues I will
address in the next section: 29 (1) agential realism grounds and situates knowledge
claims in local experiences: objectivity is literally embodied; (2) agential realism
privileges neither the material nor the cultural: the apparatus of bodily production is
material-cultural, and so is agential reality; (3) agential realism entails the inter-
rogation of boundaries and critical reflexivity; and (4) agential realism underlines
the necessity of an ethics of knowing.

(1) Agential realism grounds and situates knowledge claims in local experiences:
objectivity is literally embodied.
On the one hand, feminists and other Enlightenment critics have expressed
skepticism towards objectivism, especially
180 KARENBARAD

[tlhe idea of a basic dichotomy between the subjective and objective; the conception of knowledge as
being a correct representation of what is objective; the conviction that human reason can completely free
itself of bias, prejudice, and tradition; the ideal of a universal method by which we can first secure firm
foundations of knowledge and then build the edifice of a universal science; the belief that by the power
of self-reflection we can transcend our historical context and horizon and know things as they really are
in themselves (Bernstein, 1983,36).

In the post-Kuhnian era in which we live, the arguments against objectivism have
been robust and extensive, reaching across disciplinary boundaries and out into the
world beyond the academy, so that few scholars currently find it tenable to sub-
scribe to the set of Enlightenment doctrines outlined above. Enlightenment
defenders are hard-pressed to show how objectivism can bootstrap its way out of
the murky waters of spacetime contingencies. Ironically, mainstream anti-
Enlightenment theorists, including Derrida, Foucault, and Lyotard, have ignored
crucial social markers such as gender and race in their critiques of the universal-
izing tendencies characteristic of the Enlightenment project. However, it is not only
the limitations of these critiques that have concerned feminists, but their thorough-
going rejection of the entire set of Enlightenment goals as well. Feminist theorists
have taken exception with anti-Enlightenment scholarship that abandons the pos-
sibility of positive epistemologies in their embrace of interpretationism, relativism,
and strong social constructivism. 30
Haraway's theory of situated knowledges presents a direct challenge to the ob-
jectivist "view from nowhere", the "godtrick" of infinite passive vision, and the
equally irresponsible relativist "view from everywhere", posing embodied sight -
the view from somewhere, along with the responsibility that that entails - as the
key to feminist objectivity. According to Haraway:
There is no unmediated photograph or passive camera obscura in scientific accounts of bodies and
machines; there are only highly specific visual pOSSibilities, each with a wonderfully detailed, active,
partial way of organizing worlds .... Understanding how these Visual systems work, technically, so-
cially, and psychically, ought to be a way of embodying feminist objectivity (1988, 583).

Agential realism gives us a technology of embodiment (Barad, 1996). Recall that

concepts obtain their meaning by reference to a particular apparatus marking the
placement of a constructed boundary between the "object" and the "agencies of ob-
servation". And in tum, the point of reference for objective description of phenom-
ena is "from permanent marks ... left on bodies which define the experimental
conditions." Therefore, bodies which define the experimental conditions serve as
both the endpoint and the starting point for objective accounts of our intra-actions.
In other words, objectivity is literally embodied. According to agential realism,
knowledge is always a view from somewhere - objective knowledge is situated
knowledge.

(2) Agential realism privileges neither the material nor the cultural: the apparatus
of bodily production is material-cultural, and so is agential reality.
While theoretical constructs are not to be understood as representing trans-
parently given observation-independent properties possessed by independent
material objectslbeings as they exist in isolation from all observational interactions,
 MEETING THE UNIVERSE HALFWAY 181
neither are we to interpret these constructs as artifacts of the observational process,
purely discursive gestures imprinted on the blank slate of passive matter. As Bohr
tell us:
These problems were instructively commented upon from different sides at the Solvay meeting .... On
that occasion an interesting discussion arose also about how to speak of the appearance of phenom-
ena .... The question was whether, as to the occurrence of individual effects, we should adopt a ter-
minology proposed by Dirac, that we were concerned with a choice on the part of "nature" or, as
suggested by Heisenberg, we should say that we have to do with a choice on the part of the "observer"
constructing the measuring instruments and reading their recording. Any such terminology would,
however, appear dubious since, on the one hand, it IS hardly reasonable to endow nature with volition in
the ordinary sense, while, on the other hand, it is certainly not possible for the observer to influence the
events which may appear under the conditions he [sic] has arranged. To my mind, there is no other alter-
native than to admit that, in this field of experience, we are dealing with individual phenomena and
that our possibilities of handling the measunng instruments allow us only to make a choice between the
different complementary types of phenomena we want to study (Bohr, 1949,223).31

There are three important points that we can take from this passage: (i) nature has
agency, but it does not speak itself to the patient, unobtrusive observer listening for
its cries - there is an important asymmetry with respect to agency: we do the repre-
senting, and yet (ii) nature is not a passive blank slate awaiting our inscriptions, and
(iii) to privilege the material or the discursive is to forget the inseparability that
characterizes phenomena.
As evidenced in the above quote, when Bohr and other physicists engaged in dia-
logue about quantum theory they spoke about the "choice made on the part of the
experimenter," as if the experimenter is a liberal humanist actor of individual will. 32
There is no reference to the social dimensions of scientific knowledge production.
(It is interesting to note though that Bohr does acknowledge the role of linguistic
constraints.) However, without intending any anachronistic projections, it must be
the case that material-semiotic apparatuses are fully cultural (i.e., social, linguistic,
historical, political, etc.) frameworks, not the result of individual will, since repro-
ducibility and unambiguous communication are the criteria for objectivity. That is,
scientists make meanings within specific communities, they do not do so au-
tonomously. Therefore, according to agential realism, the apparatus that is
theorized must be a multidimensional material-cultural framework.
Furthermore, agential realism provides an account of the simultaneously material
and cultural nature of the ontology of the world. Saying that something is socially
constructed doesn't mean that it isn't real - on the contrary, according to agential
realism, reality is itself material-cultural,33 There is no opposition here between
materiality and social construction: constructedness does not deny materiality. The
materiality of the body is not dissipated by its constructedness since reality is con-
stituted by the "between ", the inseparability of nature-cultural / world-word /
physical-conceptual / material-discursive. Culture does not displace or replace
nature, but neither do things exist outside of culture. Phenomena are material-
cultural be-in's. Haraway makes a similar point, I think, in designating objects as
"material-semiotic actors". She uses this term "to portray the object of knowledge
as an active, meaning-generating part of the apparatus of bodily production,
without ever implying the immediate presence of such objects .... Boundaries are
182 KARENBARAD

drawn by mapping practices; 'objects' do not preexist as such. Objects are bound-
ary projects" (Haraway, 1988, 595). In other words, the apparatus of bodily produc-
tion, qua agencies of observation, are not separable from phenomena.

(3) Agential realism entails the interrogation of boundaries and critical reflexivity.
Wholeness, according to agential realism, does not signify the dissolution of
boundaries. On the contrary, boundaries are necessary for making meanings.
Theoretical concepts are only defined within a given context, as specified by con-
structed boundaries. Wholeness is not about the prioritizing of the innocent whole
over the sum of the parts; wholeness signifies the inseparability of the material and
the cultural. Wholeness requires that delineations, differentiations, distinctions be
drawn; differentness is required of wholeness. Utopian dreams of dissolving bound-
aries are pure illusion since by definition there is no agential reality without con-
structed boundaries. There are two common ways to attempt to deny responsibility
for boundaries: (1) claim that they are natural, or (2) claim that they are arbitrary
partitionings of a holistic oneness, existing outside of human space and time. In
contrast, agential realism explicitly shows that boundaries are interested instances
of power, specific constructions, with real material consequences. There are
not only different stakes in drawing different distinctions, there are different
ontological implications.
Furthermore, boundaries are not fixed. Productive and creative tensions are set
up in consideration of different possible placements of agentially situated cuts.
Consideration of mutually exclusive intra-actions, constituting opposing shifts in
the conceptual terrain, reminds us that descriptive concepts do not refer to an
observer-independent reality, but to phenomena. In fact, descriptions reflect back
upon the specification of boundaries, since descriptions refer to phenomena and
boundaries are in phenomena (i.e., the conceptual scheme is tied to the physical ap-
paratus and the descriptions refer to the phenomenon, which by definition includes
the apparatus; therefore the description refers back to the constructed conceptual
scheme). The placement of the boundary becomes part of what is being described:
human conceptual schema are part of the quantum wholeness. Descriptions of
phenomena are reflexive, and the shifting of boundaries constitutes a meta-critique.
The acknowledgement and interrogation of context is common to many feminist
epistemologies. For example, both Longino's theory of contextual empiricism and
Harding's theory of strong objectivity call for a critical examination of background
assumptions. Harding writes:
In an important sense, our cultures have agendas and make assumptions that we as individuals cannot
easily detect. Theoretically unmediated experience. that aspect of a group's or an individual's experi-
ence in which cultural influences cannot be detected. functions as part of the evidence for sCientific
claims. Cultural agendas and assumptions are part of the background assumptions and auxiliary
hypotheses that philosophers have identified. If the goal is to make available for critical scrutiny all the
evidence marshaled for or against a scientific hypothesis, then this evidence too requires critical exam-
ination within scientific research processes (1991.149).

Agential realism includes practice within theory: theory is epistemologically and

ontologically reflexive of context. Contrary to traditional views of theory that take
 MEETING THE UNIVERSE HALFWAY 183
the actual practice of measurement to be outside of theory, and according to the
logical positivist/empiricist program which assumes that measurements trans-
parently adjudicate among theories, Bohr's philosophy-physics entails a reconcep-
tualization of science that places the discourse on science into scientific discourse.
That is, phenomena are the embodiment of cultural practices within theory. I
suspect that the reflexive implications are a root cause of Bohr's marginalization
within the physics community (see Barad, 1995 for more details).

(4) Agential realism underlines the necessity of an ethics of knowing.

According to agential realism, reality is not independent of our explorations of
it - both epistemologically and ontologically speaking. Focusing on the ontological
as well as the epistemological is crucial to intra-acting responsibly within the
world. Knowledge projects entail the drawing of boundaries, the production of phe-
nomena which are material-cultural intra-actions. That is, our constructed knowl-
edges have real material consequences. And therefore, agential realism calls for
direct accountability and responsibility. It is to remind us of this fact that the ad-
jectival form of the word "agency" modifies and specifies the form that realism
takes here, in defiance of traditional forms of realism that deny any active participa-
tion on the part of the knower. Agency is a matter of intra-acting, that is, agency is
an enactment, it is not something someone has.
We need to understand the technologies by which particular social constructions
have real material consequences. According to agential realism, the full apparatus
of bodily production must be theorized as well - the consideration of acontextual
variables will give inadequate results. Think again of the existence of wave phe-
nomena in the context of a particular apparatus of bodily production; particle phe-
nomena are tied to a mutually exclusive apparatus. Quantum physics can account
for the phenomenon that exists in a particular context if and only if the apparatus of
bodily production is included in the calculation. Agential realism provides an un-
derstanding of the possible dynamical intra-actions of nature-culture as ontological
be-in's, thus helping us to theorize the material consequences of constructing par-
ticular apparatuses of bodily production. Knowing involves denaturalizing, multi-
ply contesting and destabilizing the existing apparatus to refigure boundaries. This
will have real material consequences, so that agential realism underlines the
requirement for an ethics of knowing.

7. AGENTIAL REALISM AND SCIENCE STUDIES

The notion of complementarity, Bohr also wants to say, can be seen to arise out of the nature of our con-
sciousness of what is "other" to us, out of the unresolvable tension between content and form, between
reality and concept, and between theory and experience. Our representations of reality do not so much
involve a pnvileged mental mirroring of external reality, in which object and subject are absolutely
distant from each other, as a successful compromise between language and activity .... Yet for Bohr the
relationship between word and world is not seen as entirely relative, with the implication that our words
have no anchorage in world; instead given the nature of our conscIOusness of what is demonstrably
"other" to us, a relationship between word and world is accepted as necessarily denying complete
resolution (Honner, 1987, 103).
184 KARENBARAD

As a scientist I have been very interested in feminist science studies in part because
the scholars in this field, many of whom are scientists as well, have resisted the po-
larization often found in contemporary discussions about the nature of science as
posed by the more traditional and monodisciplinary approaches. Evelyn Keller
identifies two noncommunicating discourses about science,
... one an increasingly radical critique that fails to account for the effectiveness of science, and the other
a justification that draws confidence from that effectiveness to maintain a traditional, and essentially
unchanged, philosophy of science. What is needed is a way of thinking and talking about science that
can make sense of these two very different perspectives - that can credit the realities they each reflect
and yet account for their differences in perception (Keller, 1985, 6).

I think these tensions are quite productive and, in my opinion, Keller's challenge
marks one of the most important issues for contemporary science studies.
If the "discovery model" of science, that sees the production of scientific knowl-
edge as a one-actor show - nature at center stage with a passive audience of
observers patiently looking on - is no longer acceptable, and neither is some
extreme version of social constructivism that presents science as an arbitrary com-
pendium of power-laden rhetorical moves, then is it possible to give a detailed
understanding of the interaction of nature and culture in the production of scientific
knowledge? Agential realism provides a framework that can be useful for re-
theorizing a range of issues generated by reliance on classical epistemologies and
ontologies. In this section, I will explore the implications of agential realism for
science studies. I have in mind the following questions: How can we reconcile the
claim of science studies scholars that scientific knowledge is a socially constructed
product that is conceptually, methodologically, and epistemologically allied along
particular axes of power with both the liberatory and oppressive interventions that
are possible because of the reliability of empirically adequate scientific knowl-
edges? What, if anything, can be said about the ontology of our world through our
investigations of it? Is there a notion of realism that is consistent with the assertion
that scientific knowledge claims are culturally specific?
The scientific method, which was our Enlightenment birthright, promised to
serve as a giant distillation column, removing all cultural influences, and allowing
patient practitioners to collect the pure distillate of Truth. The transparency of
Newtonian physics to the process of measurement grew out of and helped re-
inforce this cultural milieu of objectivism that made the successes of science
unparadoxical: science works because scientists are able to obtain the facts about
the world as it exists independently of us human beings. The Enlightenment notion
of science is premised on a separation between knowing subjects and observation-
independent objects. Agential realism challenges this conceptualization of science
on epistemological and ontological grounds.
According to agential realism, scientific concepts obtain their meaning by
reference to a particular physical apparatus marking the placement of an agentially
constructed cut between the "object" and the "agencies of observation". In tum, the
point of reference for objective description of phenomena is "from permanent
marks ... left on the bodies which define the experimental conditions"34 (Bohr,
 MEETING THE UNIVERSE HALFWAY 185
1963c, 3). Therefore, bodies serve as both the endpoint and starting point for
objective accounts of our intra-actions. In other words, agential realism gives us an
embodied account of objectivity.
Scientific results are not reproducible because we are able to measure the
observer-independent properties of an independent reality. Reproducibility is pos-
sible because scientific investigations are embodied, grounded in experience, in
praxis. Reproducibility means the possibility of the reproduction of phenomena,
and phenomena are written on the "body"; phenomena are the place where matter
and meaning meet. Reproducibility of phenomena does not require or serve as
proof for access to the transcendent. "The overall force of Bohr's argument is that
we are without absolute foundation in our participation in the world, despite the
acceptance that our language works by being anchored in everyday experience of
reality" (Honner, 1987, 222). Reproducibility of phenomena is not innocent - it
depends upon the choice of some constructed cut for which the ambiguity is only
temporarily, contextually decided in such a way as to lend meaning to certain con-
cepts, at the exclusion of others. Reproducibility is not a filter for shared biases; the
apparatus of bodily production is culturally situated. The scientists marking off the
boundaries are marked by the cultural specificities of race, history, gender, lan-
guage, class, politics, etc. In stark contrast to the classical framework, there is a
sense of agency and therefore accountability. Since reproducibility is the corner-
stone of Western science, in the context presently under discussion, science has
meaning, but not in any classical sense. 35 According to agential realism, science is
movement between meanings and matter, word and world, interrogating and
redefining boundaries, a dance not behind or beyond, but in "the between ", where
knowledge and being meet.
Scientific knowledge is not an arbitrary construction independent of "what is out
there", since it is not separate from us; and given a particular set of constructed
cuts, certain descriptive concepts of science are well-defined and can be used to
achieve reproducible results. However, these results cannot be decontextualized.
Scientific theories do not tell us about an independent reality; scientific concepts are
not simple namings of discoveries of objective attributes of an independent Nature
with inherent demarcations. Scientific concepts are not innocent or unique. They
are constructs which can be used to describe "the between", rather than some inde-
pendent reality. (Why would we be interested in such a thing as an "independent
reality" anyway? We don't live in such a world.) The point is that phenomena con-
stitute reality. That is, reality itself is material-cultural. And according to agential
realism, scientific know ledges are situated knowledges describing agential reality.
My revision of an important quote by Niels Bohr goes like this: "It is wrong to
think that the task of physics is to find out how nature is. Physics concerns what we
can say about [our intra-actions within] nature." We are in reality, we must be in
our theories. In other words, scientific theories describe agential reality - which is
just what we are interested in (we don't live in a transcendent reality). For scientific
theories to be able to describe agential reality, scientific knowledge must take
material-cultural factors into account since they are in agential reality, otherwise
186 KARENBARAD

we would not expect scientific knowledge to produce empirically adequate

accounts of our intra-actions within nature. Reliability is not premised on access to
the transcendent, but on the grounding of practice within theory. (The nonclassical
epistemology and ontology have removed the paradox of the classical position
which sees the reliability of scientific theories as contingent upon objective dis-
coveries of an independent reality.) Consideration of mutually exclusive sets of
concepts produce crucial tensions and ironies which underline the point that it is the
fact that scientific knowledge is socially constructed that leads to reliable knowl-
edges about reproducible phenomena - which is just what we are interested in.
Therefore, the understanding that science as a social practice is conceptually,
methodologically, and epistemologically allied along particular axes of power can
indeed be reconciled with the fact that scientific knowledge is empirically adequate,
that it provides effective interventions which may be used towards either regressive
or liberatory purposes.
It is not that we attempt to view nature through the lens of culture with an optics
that has varying degrees of transparency or opaqueness. We do not try to fit our
theories to reality by probing the fixed boundary between nature and culture.
Phenomena constitute our ontology. And since scientific concepts can be used to
describe phenomena and phenomena are not "out there", but are material-cultural
be-in's, agential realism provides us with a form of realism that is compatible with
social constructivism. Agential realism is a form of social constructivism that is not
relativist, does not reduce knowledge to power plays or language, and does not
reject objectivity.

8. CONCLUSIONS

So, I think my problem and "our" problem is how to have simultaneously an account of radical historical
contingency for all knowledge claims and knowing subjects, a critical practice for recognizing our own
"semiotic technologies" for making meanings, and a no-nonsense commitment to faithful accounts of a
"real" world, one that can be partially shared and friendly to earth-wide projects of finite freedom, ad-
equate material abundance, modest meaning in suffering, and limited happiness (Haraway, 1991, 187).

Agential realism denies the innocence of naive realism; instead, it entails a con-
scious, critical reflexivity. Dualisms, binary oppositions, dichotomies, and other de-
marcations are not secured with natural status as Cartesian cuts which form the
foundation of all knowledge - not even in physics. The lines drawn are power-
laden epistemological moves with stakes in a given conceptual scheme. This
doesn't mean that we can't justify drawing lines, or that crafted conceptual schemes
are unusable. Just because science is exposed as being socially constructed doesn't
mean that it doesn't work. And empirical adequacy is not an argument that can be
used to silence charges of constructivism. But neither is constructivism a proof of
epistemological relativism. I have argued that reliable theories about our intra-
actions are necessarily socially constructed theories with real material con-
sequences. We need knowledge systems that are both reliable and accountable
guides to action. Agential realism creates an alternative to objectivist accounts of
 MEETING THE UNIVERSE HALFWAY 187
knowledge production that deny the situated nature of know ledges and social
constructivist accounts that do not address the effectiveness of knowledge systems.
Agential realism is not a call for feminists and others to bow down once again to
the hegemony of science in finding a new epistemology. On the contrary, agential
realism undermines the hegemony of science (though not its effectiveness).
Agential realism insists that science incorporate a reflexive critical discourse, like
all other human endeavors. Bohr argued that quantum physics, considered by many
to be the most highly esteemed field of science, requires a new framework for
understanding the role of descriptive concepts in scientific knowledge production.
The notions of wave and particle deconstruct one another, exposing the limitations
of the classical framework. There is irony, though perhaps little surprise, in the fact
that our interactions with light - oh light! that ever resilient metaphor for knowl-
edge illuminating the dark terrain of ignorance - plays a central role in under-
mining the hegemony of Newtonian physics, that bright star of the En-light-
enment, deconstructing the objective-subjective and nature-culture dualisms that
have plagued many attempts to understand the nature of scientific knowledge.
What I am proposing is not some holistic approach in which subject and object
reunite into some apolitical relativized whole, but a theory which insists on the
importance of constructed boundaries and also the necessity of interrogating and
refiguring them. The intra-action involving the subject-object problematizes
natural, pure, and innocent separations, but not in a way which reaches for the rapid
dissolution of boundaries. Boundaries are not our enemies; they are necessary for
making meanings, but this does not make them innocent. Boundaries have real
material consequences - cuts are agentially positioned and accountability is man-
datory. The shifting of boundaries often helps bring to the surface questions of
power which the powerful often try to submerge. Agential realism insists that
mutually exclusive, shifting, mUltiple positionings are necessary if the complexity
of our intra-actions are to be appreciated. 36 Multiple contestations of agentially po-
sitioned boundaries keep concepts alive, and protects them from reification and
petrification. Our goal should not be to find less false boundaries for all spacetime,
but reliable, accountable, located temporary boundaries, which we should anticipate
will quickly close in against us. Agential realism will inevitably be a casualty of its
own design, but I suggest that there is power there presently for some of our pur-
posesY Agential realism involves located or situated knowledges (Haraway, 1988),
knowledges that reject transcendental, universal, unifying master theories in favor
of understandings that are embodied and contextual.
Who are the agents in agential realism? The history of science parallels the
history of knowledge in other arenas: the powerful effectively portray their own
knowledge systems as universal, denying their own agency. Within this tradition,
agency has been an issue quite separate from authorship. Rivalries over primary
authorship are common in the history of science, but what is at stake is cleverness
and ingenuity; what is "discovered" is presumed unmarked by its "discoverer". The
claim is that the well-prepared scientist can read the universal equations of Nature
that are inscribed on G-d's blackboard: Nature has spoken. The paradox is that the
188 KARENBARAD

objects being studied are given all the agency, even and most especially when they
are seen as passive, inert things, culture-free and existing outside of human space
and time, moving aimlessly in the void. Completing this Enlightenment scenario,
are the passive human observers who are without agency. The overdetermination of
Enlightenment discourse is revealed in the juxtaposition of this mythology with the
liberal humanist story that provides man with individual will and dominion over
nature.
The nature-culture and object-subject dualisms are constructed cuts passed off
as inherent and fixed in the service of this legacy. Agential realism makes other
moves: shifting and destabilizing boundaries. Here knowledge comes from the
"between" of nature-culture, object-subject, matter-meaning. The Cartesian split
between the agencies of observation and the object is a classical illusion. Agency
cannot be designated as residing in one or the other in isolation. The observer does
not have total agency over passive matter - not any representation of reality will do -
since not any result one can think of is possible: the world "kicks back". Neither does
the object have total agency, whispering its secrets, mostly through the language of
mathematics, into the ear of the attentive scientist - knowledge is not so innocent; it
doesn't "just come out that way" all by itself. Nature is neither a blank slate for the
free-play of social inscriptions, nor some immediately present, transparently given
"thingness". Agential realism acknowledges the agency of both subjects and objects
without pretending that there is some utopian symmetrical wholesome dialogue,
outside of human representations. Science is not the product of some interaction
between two well-differentiated entities: nature and culture, since it flies in the face of
any matter-meaning dichotomy, like an electron that tunnels through boundaries set
up to confine its motion. Meaning and matter are more like interacting excitations of
non-linear fields - a dynamic, shifting dance we call science. 38
Phenomena are the intra-actions of knowledge and being, word and world, culture
and nature. Phenomena are material-cultural be-in's. Agential realism relies on a
non-classical ontology. The material is not fixed and prior to discursive signification,
but in it. Jeanette Winterson writes in her recent novel Written on the Body: "That is
how I know you. You are what I know" (Winterson, 1992, 120). Intra-acting is an
activity that theorizes the mechanics of an embodied objectivity. In our attempt to
understand we actively participate within reality. Realism is not about representa-
tions of an independent reality, but about the real consequences, interventions,
creative possibilities, and responsibilities of intra-acting within the world.
Finally, materiality matters: there are social and material reasons for knowledge
claims - the intra-actions of the material and the discursive are the technologies of
embodied objectivity - and socially constructed know ledges have real material con-
sequences. These conceptions of materiality are opposed to the immediacy of
matter in naive realist accounts and its neglect in some social constructivist
accounts. It seems to me that giving up on realism would be as hasty as giving up
on objectivity. Feminists have interrogated, redefined, and retheorized objectivity;
agential realism is an attempt to formulate a feminist notion of realism. Agential
realism goes beyond the recognition that there are material and cultural reasons for
knowledge claims, beyond the reconceptualization of description in knowledge
 MEETING THE UNIVERSE HALFWAY 189
systems, to providing us with a positive sense of the ontology of our world and
some important clues as to how to intra-act responsibly and productively within it.
Judy Grahn suggests that: "To understand, to get to the basis, the root or hidden
meaning, is the wrong tool to bring" to our own work. "Perhaps interstand [or
better yet intra-stand] is what we do, to engage with the work, to mix with it in an
active engagement, rather than 'figuring it out'. Figure it in" (Grahn, 1989, 39).
Knowledges are not innocent representations, but intra-actions of natures-<:ultures:
knowledge is about meeting the universe halfway.

Pomona College,
Claremont, CA

ACKNOWLEDGEMENTS

The original version was presented at the Pew Gender and Science Workshop, Lake
Arrowhead, CA, September 1991. Successive versions were presented at numerous
conferences and public lectures. I am grateful to Ruth Wilson Gilmore, Donna
Haraway, Sandra Harding, Martin Krieger, Helen Longino, Lynn Hankinson
Nelson, Jennifer Rycenga, Sharon Traweek, and Roanne Wilson for their insightful
comments and support.

NOTES

I A less obvious point is that the success of scientific theories is not automatic for realists either, as

Laudan (1981) and Fine (1984) argue.

2 Cushing asserts that "realism is in double jeopardy", in the sense that although Bohm's mterpretation

IS realist, he tags Bohr's interpretation of quantum physics as antirealist, and furthermore, the existence
of thIS concrete example of underdetermination means that it would be very difficult to make the case
for realism. Although I will be argumg here for a realist interpretation on the part of Bohr. this diver-
gence in and of itself does not weaken the underdeterminatlOn aspect of Cushmg's argument. (There are
a few mdependent issues however. One is the fact that the empirIcal eqUIvalence of these theories
depends upon the resolutIOn of the measurement problem for the Copenhagen interpretation (see fn. 6)
since rigorously speaking without such a resolutIOn the Copenhagen interpretatIOn does not offer definite
predictions (see Albert. 1992). And of course. it still remains to be seen whether Bohm's theory and the
Copenhagen theory are empirically coinCIdent in all respects.)
3 This is not a cIrcularIty. As I will explain later, it is indIcative of a critical reflexiVIty .

.. The destabilization of lIberal humamst conceptIOns of identity that follow from the framework of
agential realIsm will not be my focus here. My focus here will be primarily on science. For more details
on agential realism and identity see Barad, forthcoming.
5 The collected volumes of Bohr's writmgs have been made available thanks to Rosenfeld, (1972- ).
6 While physicists talk of the Copenhagen interpretation, in one sense there are really many
Copenhagen InterpretatIOns. or to put It another way there IS no well-defined, coherent, and complete
Copenhagen interpretation. This is due to the fact that the physicists who are seen as the contrIbutors had
strong philosophical/interpretative differences, so that what is taken to be the Copenhagen interpretatIOn
is actually a superpositIOn of the views of Bohr (complementarity), Heisenberg (uncertamty), Born
(probability), and Von Neumann (collapse), to name a few of the key players.
7 For more details see Cushing, 1994. (Although Bohr's philosophy-physics is not a primary focus for

Cushmg, I note that my reading of Bohr diverges substantially from Cushing's. As I speCify in more
190 KARENBARAD
detail later, my reading has much more overlap with interpretations presented by a number of Bohr
scholars.)
8 There are an increasing number of quantum textbooks that do not mention any of Bohr's contributions

to the field (except for reference to his pre-quantum theory atomic model). That is, there is often no
mention of his principle of correspondence and the role it played in the development of the quantum
theory, or Complementarity and its importance to an understandIng of quantum theory.
9 For Bohr, Complementary means simultaneously necessary and mutually exclusive (as explained in
detail in the next section. NB: I capitalize 'complementary' when it is used in Bohr's sense of the term).
See Bohr (I 963b ) for examples for this approach. An attempt by Bohr to resolve the vitalism-mecha-
nism debate in biology failed because he assumed, from his limited technological perspective, that the
conditIons for examimng the underlying mechanics of life processes and the conditions for maintaining
the life of the specimen under investigation were mutually exclusive.
iO Although one is free to give antirealist interpretation of Newtonian physics, the "classical realist" one

articulated here is particularly seductive to our Enlightenment Intuitions, and I have heard variations of
this classical realist tenet espoused tIme and again to students in undergraduate physics classes. (It is of
course iromc to attribute a realist stance towards the phYSICS of one who was unwilling to feign any
hypothesis, but not many students would pick up on this since physics courses overwhelmingly lack any
overt discussion of the different interpretative stances with regard to science. See Barad (1995) for the
pedegogical implications of thiS widespread inattention to metatheoretIcal issues and the lack of critical
reflexivity.)
II A re-visioning of the nature of light is common to both of the major conceptual revolutIOn of

twentieth century physics: special relativity and quantum theory. Special relativity will not be con-
sidered in this paper, but a few words to distingUish some of the more popular implications of this theory
from quantum mechanics may be helpful to some readers. The special theory of relavity is based on the
empirically verified invariance and finiteness of the speed of light (llspeed of light = l/constant =
lie #- 0). Einstein transformed Galilean relatIvity into a new theory of relativity, redesignating certain
previously held invariant quantities as relalive and vice versa. (Einstein had thought of calling this
theory "the theory of invariances", and he may have been better off doing so given the political climate
in Europe during the first half of the 20th century.) The theory of relativity gives the measurement
process some limited viSibility: concepts such as "time" and "length" are defined relalive to a particular
frame of reference In which the measurements are performed. (It is not that time appears to slow down;
time is what you measure with a clock.) The theory of relativity may have undermIned the universality
of certain concepts, but the assumption that measurements are continuous and determinable is never
questioned. That is, according to the speCial theory of relativity there still IS a well-defined separation of
object and measuring instrument, i.e., a clear subject/object distinction is preserved. The properties
measured are attributable to an independent object as measured relative to a particular frame of
reference. (The frame of reference simply specifies which "time" we are talking about, that is, what we
mean by "time" in each case.)
12 It is Important to note that the fact that Newtoman phYSICS "works" in the macroscopic domain does

not mean that the assumptions of measurement transparency are true in that domain. On the contrary,
this simply explains why the assumptions lay hidden for centuries. That IS, the fact that Newtonian
phYSICS makes predictions that are approximately the same as those made by quantum theory in the
macroscopic domain is due to the fact that in that regime the ratio of Planck's constant to the mass of the
particle is smaller than the accuracy required of the macroscopic situation in question - but it IS not zero.
ThiS is why Bohr refers to the general epistemological lessons of quantum theory.
13 A rough, intuitive picture is the follOWIng: think of catching a ball, the relalive amount by which your

arm moves back IS an indication of the momentum of the ball.

14 I have chosen to use the term 'constructed' Instead of Bohr's term 'arbitrary' for two reasons. First of

all, 'arbitrary' is misleading since the cut is not totally arbitrary in that the cut must be made in such a
way that the measuring device is always macroscopIc (thiS is necessary since the use of classical con-
cepts is predicated on a subject/object split). Secondly, the term 'arbitrary' carries misleadIng con-
notations such as the inappropriate associations of relativism. The point that I think Bohr IS out to
emphasize in using the term 'arbitrary' is that since there is no inherent/Cartesian distinction some non-
inherent distinction still must be drawn. SInce the choice of a conceptual apparatus necessarily draws
 MEETING THE UNIVERSE HALFWAY 191
this distinction, I will use the term 'constructed' in the hope that this term will connote agency. The full
contrast is that classical physics is premised on an inherent/naturally occurring/fixediuniversallCartesian
distinction while quantum physics requires constructedlagentially positionedimovable/locall"Bohrian"
cuts, positioning Bohr as Descartes counterpart.
15 Another way of expressing this quantum quandary is by noticing that this means that the act of mea-
suring can be accounted for only if the measuring device IS itself treated as an object, defying its purpose
as a measuring instrument.
16 See Barad (1995) for a more detailed discussion of this example.
17 "Bohr's provocative tendency, especially in earlier writings, to 'emphasize the subjective character of

all experience' (Bohr, 1963a, I) brought his entire Interpretation of quantum theory into peril" (quoted
in Honner, 1987, 65), parallel to terminological choices made by some science studies scholars early on
that proved equally rhetorically disadvantageous.
18 Bohr interchanges the phrases "language of everyday experience" and "language of classical

physics". The connection for Bohr is that everyday experiences take place within the macroscopic
realm to which the language of claSSical phYSICS is applied. However, he does suggest that the lan-
guage of everyday experience may include "suitable generalizations" of the language of classical
physics.
19 Bohr: "It is just arguments of this kind which recall the Impossibility of subdividing quantum

phenomena and reveal the ambiguity in ascribing customary physical attributes to atomic objects"
(Bohr, 1963b, 51).
20 This detailed mapping of the relationship between classical concepts and phenomena is meant to

clarify Bohr's position with respect to the necessity of using classical concepts in the description of
quantum phenomena. Confusion about this issue is widespread in the literature. Many physicists trying
to understand Bohr's interpretative framework have accused Bohr of conservatism with respect to the
future development of physical theories: why they asked should we limit our descriptive concepts in this
way? It is also not uncommon to find philosophers describing this aspect of Bohr's theory as Kantian. I
hope that I have clearly communicated here why I think it is that Bohr was not denying the possibility
for future creative developments in physics, nor was he advocating transcendental idealism in his insis-
tence on the use of empirically grounded classical concepts even within a nonclassical framework
(which is already admitting the possibility of the evolution of Ideas). An important related fact is that
Bohr offers the observation that "everyday" languages are based on subject-predicate forms (a point that
he unfortunately makes without qualification); that is, everyday languages structurally assume
subject-object distinctions. I believe that this a contributing factor to what is commonly described as the
obscurity of Bohr's writings, since he uses many circumlocutIOns to try to talk about things that are not
inherently structured along this distinction.
21 See Keller's "Anomaly of a Woman in Physics" (1977) for one telling of a graduate school experi-
ence in the US that is typical in its discouragement of reflexiVity and contemplation of interpretative
questIons in physics, though specific in its gendering.
22 This quote is from Petersen, 1985, 302. Petersen goes on to say that Bohr had no use for an ontology.
Perhaps Bohr didn't feel the need to articulate one, but this is not to say that he held a thorough-gOing
pragmatIc or positivist view. In fact, I will argue later In this section that Bohr had a realIstIc attitude
towards wave-particle duality, for example, though his views diverged dramatically from classical
realism. Honner (1987) also argues for a realistic interpretation and against pragmatic or positivist per-
spectives, although the version of realism that Honner ascribes to Bohr does not address the issue of a
reference for our representations. Folse (1985) also advocates for an interpretation which sees Bohr as a
realist, but Folse seems to take phenomena as the result of an underlying reality.
23 Einstein et al., 1935 and Bohr, 1935. In an article entitled "Discussion with Einstein on
Epistemological Problems in Atomic Physics", for a volume honoring the epoch-making contributions
of his long time friend Albert Einstein, Bohr quotes extensively from this particularly Important passage
of his 1935 paper (see Bohr, 1949,234).
24 I have presented what may seem like a pedantic analysis of Bohr's use of the term 'phenomenon' in

this passage, but I do so because as of 1935 his use of this term was still somewhat inconsistent, and
it is therefore crucial to justify by the context of his usage of this term that it indeed is consistent
with the specific signification he assigns to it in his later writings. In fact, Bohr's usage of 'phenomenon'
192 KARENBARAD

to signify the wholeness in the interaction between "objects of investigation" and "agencies of
observation" is consistent throughout this particular 1935 article.
25 This fact motivates my introduction of the term 'intra-action' at the end of this section since phenom-

ena are the instantiation of intra-actions.

26 This criterion would apply as well to any suggestion of a Bohrian notion of anti-realism. In particular,

realists cannot expect to rely on an independent external reality, but also antirealists would be hard
pressed to argue against realIsm on the basis of some postulated inaccessible independent reality.
n This mutual exclusivity highlights the problematics of an instrumentalist stance for Bohr. How does
the instrumentalist account for the non-arbitratriness of this feature? (What auxiliary criterion must be
applied?)
28 After a few weeks of intensive discussion, Heisenberg finally acqUIesced to Bohr's point of view and

added a postscript to his article on the uncertainty principle in which he states: "In this connection Bohr
pointed out to me that I have overlooked essential points in some of the discussion of this work. Above
all the uncertainty in the observation does not depend exclusively on the occurrence of discontinuities,
but is directly connected with the necessity of doing justice simultaneously to the different experimental
data which are expressed in the corpuscular theory on the one hand and the wave theory on the other
[i.e., wave-particle duality]" (quoted in Murdoch, 1987,51) Recent papers m quantum optics (e.g., see
Scully et al., 1991) give empirical evidence in support of Bohr's interpretation of the uncertainty prin-
ciple over the one given by Heisenberg which is not consistent with these findings. NB: it IS
Heisenberg's analysis (without Bohr's corrections) that is taught to physics students. See Barad (1995)
for more details. The divergence of Bohr's and Heisenberg's interpretations of the uncertainty principle
highlights their philosophical (realist and instrumentalist, respectively) differences. The construction of
scientific theories is influenced by philosophical attitudes.
29 Other pivotal aspects of the framework of agential realism are developed in Barad, forthcoming. In

particular, there is a more in-depth discussion of the issues of agency and identity. The fact that agential
realism can be used to think about rather disparate issues from the destabilization of identity to the
destabilization of science is not a matter of a more parallelism, but different instances of the same
epistemological and ontological issues.
30 For a more detailed discussion see Harding (1990) and other articles in FeminismiPostmodernism,
ed. Nicholson.
31 The positions that Heisenberg and Dirac articulate here are consistent with the former's

instrumentalist leanings and the latter's traditional realist leanings.

32 In a related fashion I have stayed away from Bohr's term "Complementarity" because of the
associated connotations of liberal humanist conceptions of choice. Take the example of the com-
plementary (intended here in the colloquial sense of the word) theory of genders where essentialized dif-
ferences between men and women are theorized on a level playing field, denying the unequal power
relations represented by unequal material conditions. Matrix theory (not the mathematical kind but the
kmd that comes from social theory) and other nonessentializing analytic moves fully deconstruct such
liberal conceptualizations.
33 To assert that we only get to study nature through the distortmg lense of culture is to remstate the

privileged position of the transcendent once again, resulting in further claSSical epistemological
astigmatism.
34 Bohr makes direct note of this point himself: "the description of atomic phenomena has in these

respects a perfectly objective character, in the sense that no explicit reference is made to any individual
observer and that therefore ... no ambiguity is involved in the communication of information" (Bohr,
1963c, 3).
35 Notice that experiments in some fields, like high energy physics, are rarely repeated due to constraints
imposed by lImited resources or other community priorities, but the Issue here is not whether or not the
results have actually been reproduced, the issue is the possibility of reproducibility due to the literal
embodiment of objectivity. Also, note that reproducibility is still an issue for scientists studying chaotic
systems (which are highly sensitive to initial conditions) in the sense that chaotic systems do not behave
differently for different observers (it is just that it is very difficult, to start an experiment with the very
same initial conditions, but simulations of chaotiC systems are often reproduced). Other criteria delineat-
 MEETING THE UNIVERSE HALFWAY 193
ing science (read "Western science") from nonscience have been offered (see Harding, 1993). This
project of delineation is of course part and parcel of Western imperialism's focus on distinguishlOg "us"
from "them". Nonetheless, thIS same dIstinction along these lines is extremely common in the science
studies literature and it is therefore useful in this context.
36 Kondo (1989) and Sandoval (1991) make a similar pomt. Anzaldua (1987) theorizes the constructed

nature of boundaries.
37 Feminist scientists, economIsts, political scientIsts, historians, psychologIsts, geographers and literary
critics are among those who have expressed seeing the utility of agential realism for theIr projects.
38 "Tunneling" is a quantum phenomenon whereby classically confined particles escape. This is the
result of the uncertainty principle and it explains many different physical phenomena such as nuclear
decay, transistors, etc.

REFERENCES

Albert, DaVId Z.: 1992, Quantum Mechanics and Experience. Cambridge, MA: Harvard University
Press.
Anzaldua, Gloria: 1987, Borderlands/La Frontera. San Francisco: Spinsters/Aunt Lute Book Co.
Barad, Karen: 1995, 'A Feminist Approach to Teaching Quantum Physics', in Teaching the Majority:
Breaking the Gender Barrier In Science. Mathematics, and Engineering, edIted by Sue V. Rosser.
NY: Teachers College Press.
- - : 1996, 'GettlOg Real: Technologies of Embodiment and Materiality', keynote address, Gender,
Technology, Place Conference, Rutgers University, March 30, 1996.
- - : forthcomlOg, Meeting the Universe Halfway, Book Manuscript.
Bernstein, Richard J.: 1983, Beyond Objectivism and Relativism: Science. Hermeneutics. and Praxis.
Phila: U. of Penn. Press.
Bohm, David.: 1985, 'On Bohr's Views Concern 109 Quantum Theory'. in Niels Bohr: A Centenary
Volume, ed. French, A. P. and Kennedy, P. J. Cambridge. MA: Harvard U. Press.
Bohr, Niels: 1935, 'Can Quantum-Mechamcal Description of Physical Reality Be Considered
Complete?', Physical Review, 48, 696-702.
- - : 1937, 'Causality and Complementarity', Philosophy of Science, 4.
- - : 1949, 'Discussion WIth Einstein on Epistemological Problems in Atomic Physics', in Albert
Einstein: Philosopher-Scientist, ed. P. A. Schilpp. Evanston: Northwestern U. Press.
- - : 1963a, The Philosophical Writings of Niels Bohr, Vol. I: Atomic Theory and the DeSCription of
Nature. Woodbridge, Conn: Ox Bow Press.
- - : 1963b, The Philosophical Writings Of Niels Bohr. Vol. II: Essays 1933-1957 on Atomic Physics
and Human Knowledge. Woodbridge, Conn: Ox Bow Press.
- - : 1963c, The Philosophical Writings of Niels Bohr. Vol. III: Essays 1958-1962 on Atomic Physics
and Human Knowledge. Woodbridge, Conn: Ox Bow Press.
Cartwnght, Nancy: 1983. How the Laws of Physics Lie. New York: Clarendon Press.
Cushing, James T.: 1994. Quantum Mechanics: Historical Contingency and the Copenhagen Hegemony.
Chicago: University of Chicago Press.
EinstelO, A., Podolsky, B., and Rosen, N.: 1935, 'Can Quantum-Mechanical Descnption of Physical
Reality Be Considered Complete?' Physical Review, 47. 777-780.
Feyerabend, P. K.: 1962, 'Problems in MIcrophysics', in Frontiers of Science and Philosophy,
ed. Robert G. Colodny. Pittsburgh: U. of Pitt.
Fine, Arthur: 1984, 'The Natural OntologIcal Attitude', in Scientific Realism, ed. by J. Lephn. Berkeley:
University of California Press.
Folse, Henry: 1985, The Philosophy of Niels Bohr: The Framework of Complementarity. NY:
North-Holland.
Fulton, Ahce: 1990, Powers of Congress. Boston: DaVId R. Godine.
Galison, Peter: 1987, How Experiments End. Chicago: The University of Chicago Press.
Grahn, Judy: 1989, Really Reading Gertrude Stein, Freedom, CA: Crossing Press.
Hacking, Ian: 1982, 'Expenmentation and ScientIfic Realism', Philosophical Topics, 13,71--87.
194 KARENBARAD
Haraway, Donna: 1985, 'Manifesto for Cyborgs: Science, Technology, and Socialist Feminism in the
1980s', Socialist Review, 80, 65-108, (reprinted in Haraway, 1991).
- - : 1988, 'Situated Knowledges: The Science Question in Feminism as a Site of Discourse on the
Privilege of Partial Perspective', Feminist Studies, 14(3),575-599, (reprinted in Haraway, 1991).
- - : 1991, Simians, Cyborgs, and Women: The Reinvention of Nature. NY: Routledge Press.
Harding, Sandra: 1990, 'Feminism, Science, and the Anti-Enlightenment Critiques', in
FeminismlPostmodemism, ed. Linda J. Nicholson. NY: Routledge.
- - : 1991, Whose Science? Whose Knowledge?: Thinking from Women's Lives. Ithaca: Cornell
University Press.
- - (editor): 1993, The "Racial" Economy of Science: Towards A Democratic Future. Bloomington:
Indiana University Press.
Hayles, N. Katherine: 1993, 'Constrained Constructivism: Locating Scientific Inquiry in the Theater of
Representation', in Realism and Representation: Essays on the Problem of Realism in Relation
to Science, Literature, and Culture, ed. by George Levin. Madison: The University of Wisconsin
Press.
Honner, John: 1987, The Description of Nature: Niels Bohr and the Philosophy of Quantum Physics.
Oxford: Clarendon Press.
Hooker, Clifford A.: 1972, 'The Nature of Quantum Mechanical Reality', in Paradigms and Paradoxes,
ed. R. G. Colodny, Pittsburgh: U. of Pittsburgh Press.
Keller, Evelyn Fox: 1977, 'The Anomaly of a Woman in Physics', in Working It Out, ed. Ruddick and
Daniels.
- - : 1985, ReflectIOns on Gender and Science. New Haven: Yale U. Press.
Kondo, Dorinne: 1989, Crafting Selves: Power, Gender, and Discourses of Identity in a Japanese
Workplace. Chicago: Chicago Univ. Press.
Latour, Bruno: 1993, We Have Never Been Modem. Cambridge, MA: Harvard University Press.
Laudan, Larry: 1981, 'A Confutation of Convergent Realism', Philosophy of Science, 48, 19-48.
Longino, Helen: 1990, Science as Social Knowledge. Princeton: Princeton Univ. Press.
Murdoch, Dugald: 1987, Niels Bohr's Philosophy of Physics. NY: Cambridge U. Press.
Pais, Abraham: 1982, 'Subtle is the Lord... ': The Science and Life of Albert Einstein. NY: Oxford Umv.
Press.
Petersen, Aage: 1985, 'The Philosophy of Niels Bohr', in Niels Bohr: A Centenary Volume, ed. French,
A. P., and Kennedy, P. J. Cambridge, MA: Harvard U. Press.
Pickering, Andrew: 1984, Constructing Quarks: A Sociological History of Particle Physics. Chicago:
University of Chicago Press.
- - : 1994, 'After Representation: Science Studies in the Performative Idiom'. Talk given at the
Philosophy of Science Association Annual Meeting New Orleans, 14-16 October, 1994.
Rosenfeld, Leon (general ed.): 1972- ,Collected Works. Amsterdam: North-Holland.
Sandoval, Cheyla: 1991, 'U.S. Third World Feminism: The Theory and Method of Oppositional
Consciousness In the Postmodem World.' Genders, 10.
Scully, Marian 0., Englert, Berthold-Georg, and Walther, Herbert: 1991, 'Quantum Opical Tests of
Complementarity', Nature, 351, 111-116.
Traweek, Sharon: 1988, Beamtimes and Lifetimes: The World of High Energy Physicists. Cambridge,
MA: Harvard U. Press.
Winterson, Jeanette: 1992, Written on the Body. NY: Vintage Books.