Tag Archives: philosophy of science

What to do if you think researchers have overlooked a significant issue? (Day 5 of Learning Road Trip)

“What to do if we think that researchers have overlooked a significant issue for 100 years?
The case of quantitative genetics and underlying heterogeneity”

Presentation to History & Philosophy of Science Department at Indiana University, with publicity to science departments
Continue reading

Social Constructions IV—Process In Time Vs. Correspondence With Process Backgrounded


The preceding discussion of co-construction vs. separate things being constructed has, at many points, drawn on science being seen as a process in time.  Let me explore further the implications of this perspective.

In describing his second constructivism, the one closest to his basic realist sensibility, Sismondo employs the image of construction of a geometric proof.  Geometric construction, however, involves not just a static arrangement of fixed points (as Sismondo describes the idea), but steps, each building on the previous ones, to achieve the proof.  When we downplay the associations construction has with a process over time we are more easily pressed to answer the static question of what, ‘after all is said and done,’ scientific knowledge corresponds to.  Sismondo, for example, in supporting some version of realism, argues that, if theoretical ‘assumptions were not sometimes approximately true [did not map reality] then it would be extremely difficult to understand how scientists achieve the pragmatic successes they do.’[i]   He considers this a strong argument by itself (as do leading realist philosophers[ii]).  From the perspective that science is co-constructed, however, the argument lacks a crucial component: what the practices of agents in real time are through which (approximate) truth leads to the pragmatic success (or lack of truth to failure).[iii]  In contrast, studies of co-constructional processes generally draw attention to the diverse practices of agents in the process of science in the making, and avoid tracing pragmatic success back to truth.  So, although some established theoretical assumptions could be true, the contribution of any of these assumptions to the process of co-construction is difficult to separate; it remains contingent on how the other components are linked in the production of the success.

When philosophers do address the issue of process, they usually invoke or imply some evolutionary scheme in which ideas that map reality best will best survive through experimental tests and disputes over correct interpretations.  But what do scientists do in this scheme — surely not vary their ideas randomly like genetic mutations?  What are the processes through which agents can bring about this “survival of the realest?”  Without such details evolutionary schemes tend to collapse to a tautology of conceiving realest as those surviving at any given point of time, or, at best, the schemes have to stress the current function (realness) of ideas, as if a history of the ideas surviving because they were realer can be inferred from the current function alone.

Philosophers are by no means alone in backgrounding process or relying on notions of correspondence.  A formulation of social constructivism, more common in talk than writing, holds that if knowledge is not given by Nature, it must be given by Society — or, with similar effect, by historical context, class structure, social location, social interests, ‘form of life,’[iv] or membership in a ‘relevant social group.’[v]   Construction evokes an image in which Society, external to science, determines, penetrates, or is reflected in the content of accepted scientific theories.[vi]  The resulting science then corresponds to the Society in which it is generated or accepted.  Admittedly, most published work is more subtle.  The literature generally presents the society-science relationship as refracted, allowing for the observation that not all of social group X believe Y and not all believers of Y come from social group X.  Scientists (and others) should not, in this view, be seen as ciphers for society or dupes for interests.  Instead, for example, they are described as producing and judging knowledge according to how it furthers goals (over and above establishing knowledge) of their social group (Shapin’s “instrumental model” of sociology/ sociological history of scientific knowledge[vii]).  While process and practice seem to be in the foreground, construction in the simpler correspondence sense has not been banished in such accounts.  If one asks how they explain why this knowledge was accepted and not that, and how this knowledge was generated in the first place, the implicit explanatory structure is more often than not one of correspondence between knowledge and interests.[viii]


The directions co-constructivism point us may not be to everyone’s liking.  For many scholars reduction of complexity and some backgrounding of on-going process would seem to be necessary if they are to say anything clear, systematic, general or useful about science.  Although I recognise that there is a lot more work to be done defining, developing and establishing co-constructivism than this note could accomplish, I disagree with the assumption or pre-judgement that such a project is unworkable.  In fact, the challenges of co-constructivism seem difficult to avoid once we ask the question: What does it mean practically for agents to modify scientific activity?  The terms just highlighted conjure an image of construction as a process of agents building  by combining a diversity of components (as in people building or remodelling a house).  So, given that the question captures in very broad outline the project of science studies, let us highlight the building sense of construction and weed out the persistent idea that science reflects or corresponds to something.  Any scientific product is part of a complex achievement; science as it is being made is being co-constructed.  In this light, the metaphor of construction can yet be productive of theory, method, challenging questions, and new perspectives on long-standing debates.

[i]  S2, p. 565

[ii]  R. Boyd, ‘On the current status of scientific realism’, in R. Boyd, P. Gasper, and J. D. Trout (ed.), The philosophy of science  (Cambridge, MA: MIT Press, 1991), 195-222.  See p. 207.

[iii]  See D. Hull, Science as a process: An evolutionary account of the social and conceptual development of science Chicago: University of Chicago Press, 1988).  For a relevant critique of natural selective explanations even in biology see P. Taylor, ‘Historical versus selectionist explanations in evolutionary biology’, Cladistics , Vol. 3 No. 2 (1987), 1-13.

[iv]  H. M. Collins and T. J. Pinch, Frames of meaning: The social construction of extraordinary science . (London: Routledge & Kegan Paul, 1982).

[v]  T. Pinch and W. Bijker, ‘The social construction of facts and artefacts; or how the sociology of science and the sociology of technology might benefit each other’, Social Studies of Science , Vol. 14 (1984), 399-441.  See also P. Rosen, ‘The social construction of mountain bikes’, Social Studies of Science , Vol. 23 (1993), 479-513.

[vi]   See the diagrams in N. Wise, ‘Mediating machines’, Science in Context , Vol. 2 No. 1 (1988), 77-113.

[vii]   S. Shapin, ‘History of science and its sociological reconstructions’, History of Science , Vol. xx (1982), 157-211.

[viii]   S. Woolgar, ‘Interests and explanation in the social study of science’, Social Studies of Science , Vol. 11 (1981), 365-394; A. Pickering, ‘The mangle of practice: Agency and emergence in the sociology of science’, American Journal of Sociology , Vol. 99 No. 3 (1993), 559-589

Social Constructions III — Co-Construction Vs. Separateness

(Continued from previous post) We can consider how Sismondo’s philosophical and political arguments are limited by his inattention to co-construction and process.


While Knorr Cetina may agree with Sismondo on some issues, she strongly opposes him when she insists that a phenomenon does not meaningfully exist, in the sense of being ‘out there,’ until we represent or reach closure about it.  Explanation seems to be the issue here; Knorr Cetina’s definition of existence has the effect that we cannot invoke pre-existing reality when explaining what happened as science was being made.  Excluding unobservables from our accounts is a strong explanatory stipulation in any science, one very difficult to maintain consistently.  Sismondo’s island analogy gently reminds us of the explanatory economy we can derive from invoking something having causal effects before any person had represented (or misrepresented) them.  The reason Knorr Cetina favours Sismondo’s rejected fourth constructivism is, I suspect, defensive.  Give unmediated reality an inch and realist philosophers will take a mile; unless she banishes unmediated reality from explanations, everything sociological becomes vulnerable to being discounted as secondary distortion of an underlying state of affairs, namely that we have ended up with theories that get it right (approximately) about how nature works.  Ironically, however, her move to insulate sociology from being so discounted by philosophers makes their work easier.  All they have to do is convince themselves that existence without representation is, in fact, meaningful, that is, one can, in principle, invoke it in explanations.  Then they can relax and continue to discount sociological challenges to philosophy.

Attention to the process of science in the making as a co-construction, however, has the virtues of allowing us both to admit unobservables into explanations and not to let philosophers off the hook so easily.  Let me explain this by extending Sismondo’s own uninhabited Pacific island analogy.  The existence of the island before we encountered it does not by itself explain how we found it, nor tell us how to find others.  And the fact that now the island can be located in many ways and by many people other than those involved in the original finding does not warrant giving its existence special status in understanding the original finding.[i]  I am not claiming people can individually or collectively wish an island into or out of existence, but little follows from this concession to unmediated reality by itself.  The claim to have found an island might be vulnerable if the island were not there, but this would not automatically be the case.  If would-be debunkers decided to revisit the contested point on the globe, they would need to get support to launch their expedition  — no trivial exercise.  They might have to cultivate patrons, lay in stores of food, procure maps, calibrate chronometers, and so on.  Their efforts in any one of these areas would influence efforts in the others, either facilitating or hindering them.[ii]   The realist (about the existence of islands) could give them little advice about these stages of the debunking process, save telling them, for the sake of credibility, to leave behind hallucinogenic drugs.

Leaving the analogy to return to the co-construction of science, I am quite prepared to believe that some deep underlying unmediated reality is fixed and to concede that social resources are only firm at best.  This does not, however, warrant a realist philosophy  Instead we should abandon the dichotomy of realist vs. relativist (idealist, conventionalist) explanations, because, when an outcome is the result of diverse fixed points and resources being linked and built upon, it is difficult to partition relative importance among the different contributing points and resources.  In other words, as science is being made, the importance of something, whether fixed, firm or malleable, is a function of the other things with which it is being linked.[iii]

Co-construction also undermines Sismondo’s political justification for maintaining a realist position.  I agree that incorrect scientific knowledge may inform social practices that we, as social critics, may want to change, and it may sometimes be effective politics for us to focus on contesting that knowledge as misrepresentation of reality.  Nevertheless, it does not follow that such critique is always necessary or even important for producing the desired social changes.  If we are to identify where and when science-centred critique could be linked into the reconstruction of the social practices we oppose, we need to understand the on-going construction of those practices.  Again, nothing follows from the truth (or from the falsity) of knowledge claims by itself.  While it may be galling that our political opponents invoke as truthful what we see as misrepresentations, there are no logical reasons either to assume that their practices are most vulnerable around those misrepresentations, to think that we can contest those misrepresentations without simultaneous attention to other contributing strands of their practices, or to fear that we weaken our politics by focussing our efforts against such other strands.

(continued in the next post)

[i]  Such a many-one relationship does, however, raise issues about whether we can generalise about science in the making and how we would do so.  In particular, we need to be able to acknowledge regularities across different makings of the science (analogous to the island’s existence being affirmed by ever more expeditions being able to locate it).

[ii]   For example, with patrons comes money, and how much money influences what food can be bought.  Similarly, potential patrons might become excited only if the promise were made to bring back slaves, requiring the expedition to plan a diversion to visit islands known to be inhabited.

[iii]  See R. C. Lewontin, ‘The analysis of variance and the analysis of causes’, American Journal of Human Genetics , Vol. 26 (1974), 400-411.)

Social Constructions II—heterogeneity and process omitted from accounts of the term (cont.)

(Continued from previous post)  Sismondo’s advice raises issues more serious than our questioning the status of philosophy over (or its separateness from) social studies of science. While he asserts that his constructivisms are separable and claims clarification will result from distinguishing a focus on social objects from a focus on the natural world, his argument, examples and footnotes suggest such separateness cannot be sustained in practice.  Again, taking each of Sismondo’s first three constructivisms (but flipping the order of 2 and 3 for expository reasons):

1) His examples of large social objects (or projects) are ‘genders, power, emotions… cities, economies, legislation and knowledge.’[xiv]   As social objects, cities and economies are clearly very material as well as being actively represented and full of meanings.  The degree to which representations of them render their material aspects more difficult to construct differently (or acrete upon differently), and vice versa, is an obvious question for investigation.  Surely knowledge is not the odd thing out in his list of examples, being the one meaningful thing (social object) that has a relationship only in the direction from material/natural to representation/ social.

3) The ubiquity of manufactured reality in laboratory science confirms that knowledge is not the exception, as Sismondo makes clear in a footnote: The ‘phenomena that science studies are extremely dependent on thoughts and theoretical commitments, for often these would not exist were it not for the experimental development of these theoretical commitments.’  The question then arises, why is it important to try to place a line to keep manufactured reality strictly separate from some deep underlying unmediated reality?

2)  In the light of Sismondo’s expanded lists of ‘fixed’ points, the answer becomes uncertain.  When these points included only data and observations one could argue that some ‘geometrically’ constructed conceptual entities come to be accepted over others because they map reality better, because they are approximately true.  But the ‘because’ is more difficult to sustain when construction builds as well upon ‘resources and the like,’ that is, upon social objects; acceptance of knowledge (or reliability of technological interventions) becomes, instead, a part of a more complex achievement.  (I develop this point further below.)

Linked together in this way, Sismondo’s conclusions invite us to subvert his own distinctions by considering the co-construction of knowledge and material reality, extending from laboratory manufactured reality to larger social projects.  It is true, as he says, that in social studies of science different things have been described as being constructed — the natural world, scientific phenomena and techniques, acceptance of facts and theories, on-going scientific activity, ‘social objects,’ or society or more generally.  At the same time, however, the literature has increasingly described practices in which these things are interlinked: scientific objects appear to be resources for people building networks to support theories; theories resources in the organisation of scientific work; language, tools, and scientific work relations resources enabling particular manufacturings of reality, and so on.  This kind of constructivism shifts perspective not just from separate things to jointly constructed sets of things, but from thinking mostly about the constructed state of the outcomes to examining the processes of their co-construction.[xv]  Of course, from the point of view of the philosophically minded, sociologists of science have provided stories about such complexity, but have yet to tease out the causal or explanatory claims implied in descriptions of such interlinkings.  An examination of the implications of such an explanatory co-constructivist project is beyond the scope of this note.  We can, however, consider how Sismondo’s philosophical and political arguments are limited by his inattention to co-construction and process.  (Continued in the next post)


[xiv]  S1, p.547

[xv]  Sismondo’s focus on the status of outcomes leads him to address Latour’s and Woolgar’s work only in terms of the fourth category of constructivism, the one he rejects.  Both Latour and Woolgar are constructivists also in the sense of co-construction (see note 8).  Cf. Sismondo’s description of Latour’s work on scientists as accumulators of resources (op. cit. note 3) as “not obviously constructivist” (S1, p. 537).

Social Constructions—heterogeneity and process omitted from accounts of the term

The Social Construction of What? (Harvard UP, 2000) by philosopher of science, Ian Hacking, critically reviews the possible meanings of social construction in the context of scientific knowledge and technology.  However, there is one meaning of construction that he does not consider, perhaps the most obvious one to the common person, namely, the process of building a structure from diverse materials, as in the foundations, frames, walls, roof, plumbing and electrical circuits, and so on.  Several years before I had raised this point in “Co-construction and process: a response to Sismondo’s classification of constructivisms,” Social Studies of Science, 25 (2): 348-359, 1995.  (My title was “Heterogeneous construction and process,” but editor insisted on substituting “co-construction” wherever I had “heterogeneous construction.”  Sismondo is another Canadian philosopher of science, at that time a student at Cornell University where I worked.)  I am not aware of other commentaries that examine this omission.  This post and the following, therefore, extract from that paper.


Any classification into types can clarify our view of the whole while, at the same time, distracting our attention from hybrids and the processes by which they are formed and sustained.[i]  In this light, the recent review by Sismondo, which teases out some of the multiple meanings given to the term ‘construction,’ and his subsequent exchange with Knorr Cetina,[ii] should leave us troubled.  Many of us are interested in the processes of science in the making, in which scientific theories, materials, tools, language, institutions, and wider social relations are being co-constructed, and are trying to analyse the diverse ‘resources’ drawn upon by agents in such co-construction processes.[iii]   Sismondo’s classification makes little space for that strand of social studies of science, focussing as it does on the type of thing being produced, not the processes of their production.  Knorr Cetina does not take issue with him on that account.  She applauds his review as an overdue clarification of constructivisms (constructionisms) and, after a brief plug for philosophers to become more sociological, centres her response on defending a conceptual claim about representations preceeding existence (more on that issue later).  If clarification means providing distinctions we should work with, we should be less satisfied with Sismondo’s taxonomy.  I feel like a misfit, and so, I suspect, do the many who have over the last decade been attracted to ideas such as ‘ecologies of knowledge,’[iv] ‘intersecting social worlds,’[v] ‘heterogeneous engineering,’[vi] and actors’ ‘networks’ of resources.[vii]   This note, however, does not criticise Sismondo just for the omission of a major category of constructivism,[viii] but argues that, from the perspective of what is omitted, his classification scheme breaks down.  The distinctions do not hold in practice and Sismondo’s conclusions about reconciling social studies of science with philosophy and about politics are not justified.


Sismondo claims that social studies of science can benefit from distinguishing four separable uses of the term construction, differing in the type of thing being produced.  The payoff derives in part from clearing up the confusions that result when different authors (or the same author in different places) are arguing from different interpretations of the term.  The rest of the benefit derives from letting go of the last of the four constructivisms, namely that things do not exist until we represent and make them meaningful.  As Sismondo interprets it, this view is metaphysically untenable.  In his analogy, while the uninhabited Pacific island is only meaningful when it is encountered and charted, it certainly existed beforehand and would not have been found otherwise, so we can now meaningfully invoke its existence-before-encounter in our explanations of its discovery.  The fourth constructivism, in contrast, implies that successful accounts of the world are unconstrained by the underlying nature of material reality.  Such relativism should also be opposed, he adds in his reply to Knorr Cetina, because it is an obstacle to politically valuable analyses of the scientific inadequacy of certain beliefs.  Then, once Sismondo has rid us of this troublesome beast, he is quite relaxed about the other three senses of constructivism he discerns in sociology of science; each can be reconciled with the realist and politically motivated philosophy of science he favours.[ix]

Consider, however, the fine print of the reconciliation that follows rejection of the fourth constructivism.  Sismondo’s three pieces of advice (corresponding to his first three constructivisms) are that we should:[xi]

1) pay attention to the contingent interaction of many agents, possibly in conflict, as they make social ‘objects’ in science (institutions, gender relations, power, and, in particular, knowledge) by acretion from previous social objects;

2) extend our notion of ‘fixed points’ from which conceptual entities are constructed (as in a geometrical proof) to include, not just ‘data and observations,’ but also ‘tools, resources, and the like’[xii]; and

3) explore the conceptual implications of science’s making extensive use of laboratory artefacts, in the production of which (unmediated) nature has been systematically excluded from the manufactured reality.

Given that these are directions already taken in sociology of science, he is, in effect, advising philosophers of science to follow sociology of science’s lead.  If this constitutes a reconciliation it is not one in which philosophy preserves its own terms, the separateness of its turf, and its status as arbiter of ‘plausible positions.’[xiii]

But Sismondo’s advice raises issues more serious than our questioning the status of philosophy over (or its separateness from) social studies of science (continued in the next post).

[i]  Of course, some hybrids disappear when the particular classification can be refined or replaced by one on a different basis.  (Orange and purple are hybrids if colours are divided into red, yellow or blue, but not if we subdivide the colour spectrum further.)  But even when re-classifying is possible, one still has to address users of the original, hybrid-entailing classification.  This note is concerned with just such a situation.

[ii]  S. Sismondo, ‘Some social constructions’, Social studies of science , Vol. 23 No. 3 (1993), 515-554 [S1]; K. Knorr-Cetina, ‘Strong constructivism – from a sociologist’s point of view: A personal addendum to Sismondo’s paper’, Social studies of science , Vol. 23 No. 3 (1993), 555-563 [KC]; S. Sismondo, ‘Response to Knorr Cetina’, Social studies of science , Vol. 23 No. 3 (1993), 563-569 [S2].

[iii]  B. Latour, We have never been modern (Cambridge, MA: Harvard University Press, 1993) discusses the challenges of hybrids for science studies and social theory, but his book does not deal with the conceptual implications of treating construction as a process.  A recent (post-Sismondo) article by A. Pickering, ‘The mangle of practice: Agency and emergence in the sociology of science’, American Journal of Sociology , Vol. 99 No. 3 (1993), 559-589, shares with this note an emphasis on process and co-construction (his ‘mangle’ and ‘impure dynamics’).

[iv]  C. Rosenberg, ‘Wood or trees?:  Ideas and actors in the history of science’, Isis , Vol. 79 (1988), 565-570.  See also S. Star, ‘Introduction: The sociology of science and technology’, Social Problems, Vol. 35 (1988), 197-205.

[v]  A. Clarke, ‘Social worlds/arenas theory as organizational theory’, in D. R. Maines (ed.), Social organization and social process: Essays in honor of Anselm Strauss  (New York: Aldine de Gruyter, 1991), 119-158.

[vi]  J. Law, ‘Technology and heterogeneous engineering: The case of Portugese expansion’, in W. E. Bijker, T. P. Hughes, and T. J. Pinch (ed.), The social construction of technological systems: New directions in the sociology and history of technology  (Cambridge, MA: MIT Press, 1987), 111-134.

[vii]  B. Latour, Science in Action:  How to follow scientists and engineers through society (Milton Keynes: Open University Press, 1987).

[viii]  We might also use the label heterogeneous constructionism, to capture an emphasis on the heterogeneity of elements or resources drawn into the practice of science in the making.  Under this general label, I would distinguish two strands, the first emphasising rhetorical, interpretive and textual tactics in securing belief, and the second centred on asking what it would mean practically for agents to modify scientific activity.  It is beyond the scope of this note, however, to develop the meaning and implications of heterogeneity, so I use the simpler label, co-construction and will not make anything of the distinction between rhetorical/textual practices and a more general sense of practice.

[ix]  In S1, note 6, Sismondo states a “minimalist” definition of realism: “scientific terms often refer to antecedently existent entities in the world,” and, by implication, our explanations of science can refer to the (pre-)existence of such entities prior to our accounts of them.  In practice, his (and other realists’) arguments use a stronger version, namely, that this fact is central to making sense of science’s successes.  It is this stronger version that I call into question here.

[xi]  S1, pp. 547-8.

[xii]  Cf. S1, pp. 516 and 547 in S1.

[xiii]  S2, p.566

Schema on debates about scientific realism

11 Sept. 1992




TOPICS         1.  Ontological 2.  Social Epistemological 1 to 2


2 to 1





1.0     There exists reality 2.0     There exist social processes of exploring “reality” = that which has been resisting all efforts at modification. R2.0  See 1.4


R1.0   We have no need to deny this, but it doesn’t come into play in exploring 2.2; see 2.4.



for this


1.1     The success of science based on this assumption and technology based on this science. 2.1          Observation of actual scientific practice


  R1.1   A science’s predictiveness or a technology’s effectiveness is contingent on more resources than reality; see 2.2.

To be



1.2      Reliable methods for discovering reality (or closer approximations to it.) 2.2     What’s rigid, what’s malleable in the process of construction of “reality” (=harnessing of cross-linked, heterogeneous resources). R2.2   See 1.45.  Also do they have anything to say about better method?


R1.2   Focusing on this gets in the way of representing well the practices of science (i.e., it’s “method talk”).






1.3     a) Truthful representation; b) In more recent accounts: theory that underwrites reliable experimental manipulations.


2.3     a) As an academic epistemological venture: better representation of the process of science; b) but, as a side product: heterogeneous recon-structions indicate multiple sites of potential intervention; c)  In a future, fully developed social epistemology: simultan-eously representation and intervention. R2.3   Are they sure this style of representation is a better basis for intervention?


R1.3   Reliable manipulations remind us of the active construction of what counts as knowledge, i.e. 2.0.  Why not start with 2.0?
  1.35  Some foundation is needed to justify interventions in the world. 2.35  Some standpoint is unavoidable for intervening in the world. R2.35  Adopting a standpoint sounds like an invitation to be dogmatic. R1.35  Displacing their stand-point away from themselves onto some form of “reality” is a particular form of 2.0 & not one we support; see 2.3.

Views on the

other side


1.4     Given that there exists reality (1.0), how can they believe that anything could be true (relativism)?  or 2.4     Why do they invoke reality when they have only observed “reality?”

Answers:  a) To maintain their position as special commentators on a privileged enterprise;  or  b) To cut through social construction and appear to have a less contingent standpoint.  (Both a & b are social constructions.)

R2.4   See 1.35


R1.4   We’re not relativists sensu “anything goes”; see 2.2.


  1.44  How can they believe that nature doesn’t play a special determining role in what counts as knowledge & what’s effective in technologies?  or     R1.44  How methodologically can they separate effects?  (Recall our starting point 2.0 and the cross-linking of resources in 2.2.)


  1.47  a) How does society get into scientific knowledge?  or

b)  How does society have a systematic effect on science?

    R1.47 a)  Society is never out of science – problems, categories, standards of evidence etc. are not given by nature.

b)  There is lots of particularity & contingency in practice.

Gender, Race, and the Complexities of Science and Technology: A bibliography

In a 2011 graduate course on “Gender, Race, and the Complexities of Science and Technology,” students were asked to add an annotated reference or resource (=person, organization…) to the evolving googledocs bibliography each week.  (Annotations were to convey the article’s key points as well as its connection to the student’s own inquiries and interests.)  The result is as follows: Continue reading

Categories in epidemiology

Idea: Collecting and analyzing data requires categories: Have we omitted relevant categories or mixed different phenomena under one label? What basis do we have for subdividing a continuum into categories? How do we ensure correct diagnosis and assignment to categories? What meaning do we intend to give to data collected in our categories?

1. We can identify a chain of steps in scientific inquiry in which each step involves assumptions and is open for negotiation and wider influences (Taylor 2005, chapter 2).

All possible phenomena

  • (-> experimental manipulation)
    • -> phenomenon deemed interesting
      • -> questions asked
        • -> categories demarcated
          • -> observations made
            • -> data collected
              • -> patterns perceived
                • -> predictions made
                  • and/or hypotheses about causes
                    • -> actions supported

Decisions made at early steps influence outcomes at later steps. E.g., if schizophrenia is used as a category as defined by the DSM, it is harder for a clinician to pay attention to the contextual and life history information of patients (Poland 2004). This is not a one-way sequence. There is also the possibility that desired outcomes for the later stages (especially the actions the researcher favors in advance and would like to be supported by the inquiry) influence decisions made at earlier steps (as indicated in the following schema).

2. When reading a study, take note of:

  • a) where the categories demarcated seem to favor certain kinds of action over others (e.g., Galton only collected data about similarities among relatives so there was no way he could explore hypotheses about non-hereditary or environmental influences or illuminate action regarding those influences); and
  • b) what kinds of remedies you would propose whenever the categories seem limited (e.g., disaggregate the category “approve of Congress,” which includes Democrats who want the Democratic majority in the Senate not to accede to fillibustering Republicans and Republicans who don’t want Democrats to get their way).

Hymowitz (2007) [not a scholarly article] disaggregates divorce rates in the USA, which hide different phenomena and trends in different social classes.

Pickles and Angold (2003) review the debate about whether categories of psychopathology are best thought of as categorical (e.g., one has schizophrenia or doesn’t) or dimensional (e.g., there are degrees of schizophrenic behavior).

Poland (picking up on both points above) argues that the category “of schizophrenia and the associated received view [does not] have anything useful to add to clinical practice concerned with severe mental illness.”

George Brown (UK) and Bruce Dohrenwend (USA) have done research for decades on the relationship between mental illness and life events or difficulties. Brown (as described by Birley and Goldberg 2000) developed methods that tried to expose the meaning of an event for the person and was critical of the US emphasis on “objective” surveys (where the same event, e.g, death of a spouse, might have very different meanings and significance for the subject). Dohrenwend describes his group’s eventual realization of this issue, but they still wanted to measure events without having the context fused into the rating of the event.

Davey-Smith et al. (2000) consider comparative methods for studying socioeconomic position and health in different ethnic communities, e.g.,  — Does socio-economic status (SES) mean the same thing for different communities? If not, what is our proposed remedy?

(This post continues a series laying out a sequence of basic ideas in thinking like epidemiologists, especially epidemiologists who pay attention to possible social influences on the development and unequal distribution of diseases and behaviors in populations [see first post in series].)


Birley, J. and D. Goldberg (2000). George Brown’s contribution to psychiatry: The effort after meaning. Where Inner and Outer Worlds Meet. T. Harris. London, Routledge: 55-60.

Brown, G. W. and T. O. Harris (1978). Sociology and the aetiology of depression; Depression and Loss; A Model of Depression; Summary and conclusions. Social Origins of Depression: a Study of Psychiatric Disorder in Women. New York, Free Press: 3-20; 233-293.

Davey-Smith, G. et al. (2000). Ethnicity, health and the meaning of socio-economic position Pp. 25-37 In Graham, H., Ed. Understanding health inequalities. Buckingham [England], Open University Press.

Dohrenwend, B. P., K. G. Raphael, et al. (1993). The structured event probe and narrative rating method for measuring stressful life events. Handbook of Stress: Theoretical and Clinical Aspects. L. Goldberg and S. Breznitz. New York, Free Press: 174-199.

Hymowitz, K. S. (2007). “Marriage and Caste in America: Separate and Unequal Families in a PostMarital Age.” Heritage Lecture #1005.

Pickles, A. and A. Angold (2003). “Natural categories or fundamental dimensions: On carving nature at the joints and the rearticulation of psychopathology.” Development and Psychopathology 15: 529-551.

Poland, J. (2004). “Bias and schizophrenia.” Pp. 149-161 in P. J. Caplan and L. Cosgrove, eds. Bias in Psychiatric Diagnosis. Lanham, MD: Rowman & Littlefield.

Taylor, P. J. (2005). Unruly Complexity: Ecology, Interpretation, Engagement. Chicago, University of Chicago Press.