28/01/2015

Science and Social Science - Differences and Similarities

Science and Social Science - Differences and Similarities
By
Rene Brauer

In this essay I intend to explore the differences and similarities between the natural and the social sciences (NSS). The essay starts with contrasting the traditional (but still popular) view of science with that of contemporary science studies. This is followed by taking this view and discussing the implication for the differences and similarities between NSS. The aspects that are addressed in regard to the differences of the NSS are the implications of having a study object rather than a study subject. Aspects of convergence that are addressed in regard to the NSS are the core assumptions, how consolidation is achieved and what the influence of the framework.

 “Das Wiederfinden dessen, was der Mensch in die Dinge gesteckt hat, heißt Wissenschaft.”[1]
Nietzsche (1844 - 1900)
What is science?

The above quote captures modernity’s conceptualization towards science. I.e. that science can ‘rediscover’ truths about the world. Only distortions of the human mind (e.g. ideology, ego or biases) blemish this process. When we keep these in check, truths about the world can be discovered. This proposition can be found in the writing of Russell, Popper and Merton and still influences the popular characterisation of science today (c.f. Sismondo 2011:1-8). This approach has been called Mode 1[2] of science studies, and from a contemporary science studies point of view (Mode 3) this conceptualisation appears to be “intellectually bankrupt” (Collins and Evans 2002:240). Mode 2 of science studies; that looked at the actual mechanisms of ‘science in action’ (e.g. Shapin, Schaffer et al. 1985; Bijker, Hughes et al. 1987; Latour 1987) came to a different conclusion. Stating that;

“There is no abstract and logical scientific method apart from evolving community norms”

Nietzsche coined the concept of perspectivism – i.e. that all ideations take place from particular perspective. Wissenschaften[3] are no exception to this proposition. This implies that consolidation of ideas, within science, cannot be achieved internally (c.f. Chalmers 2013), but rather are a product of social processes (c.f. Collins and Pinch 1993). All that exists are heuristics, which are the result of consolidated community norms. Now this is not to say that these aren’t useful, quite the contrary (c.f. Latour 2013). But this does represent a fundamental different approach from the aforementioned conceptualisation of science being a direct access toward ‘truth’ (according to Mode 1). In fact, the whole idea of ‘truth’ and how it is achieved is redefined (c.f. Latour, Harman et al. 2011). This conceptual move and its epistemological implications put the practitioners of science on the centre stage, rather than ‘truth in itself’.

The differences and similarities between science and the social sciences

A similarity arises from this conceptualization. As it implies that all different sub branches of science, despite their apparent differences, are on a fundamental level all human endeavours. Research in this view; pushes the boundaries of (socially) consolidated knowledge into areas where consensus is not yet established. Each and every sub branch has its own popular way to deal with this. In the natural sciences, the term paradigm has been used to describe this approach to how research ought to be done, according to the prevailing consensus (i.e. paradigm) of the time (c.f. Kuhn 1970).

Now despite both natural and social sciences (and humanities) representing such human endeavours there are differences between the two. One major difference is in the way actors can be treated. In the natural sciences you have a prevailing reductionist approach turning the actor into a 'passive' research object that is to be studies. Meanwhile the social sciences have more ‘active actors’ (research subjects). The difference lies in the degree of actionability. This is because the natural sciences tend to deal more with material actors, while the social sciences deal with human actors. (C.f. Law 1999; Latour 2005 on their definition of actor). This difference allows the natural sciences to manipulate the objects in such a way which wouldn't be possible with research subjects, as no protest can be voiced. This difference strengthens the power position of the researcher within the natural sciences in the research process. As such, the result  of this research process (e.g. theories) appear to have an air of certainty which cannot be achieved within the social sciences to the same degree (Camic, Gross et al. 2012:26). This (with other factors) creates a ‘hierarchy of certainty’ which usually places pure mathematics at the far end of the high certainty spectrum, while the social sciences occupy a lower position.

In a paradoxical way this makes the social sciences more ‘natural’ than the natural sciences. Because, in the natural sciences “theories don’t describe the [real world … rather they] describe either idealizations or other kinds of fundamental structures” (Sismondo 2011:166) that are supposedly representative of the real world. Due to the dissent that the subject can voice, this high degree of abstraction is not possible within the social science, hence making them more ‘natural’ (if we apply this line of reasoning). Despite, or rather because, of this difference some social scientists, departing from a view of science based on Mode 1, lament this fact. This tendency is usually characterised as physics envy[4].    

Mode 2 science studies had a heavy bias towards studying the natural sciences. This left a knowledge gap in how the social sciences operate in their actual practice (see Camic, Gross et al. 2012). However, despite this gap, some inferences can be made upon the practice of the social science, because as shown above, they also represent a human process. Three similarities are worth mentioning; the core assumption of how science ought to be done (departing from Mode 1), how it is consolidated in practice and the framework in which they operate in.

The core assumption is that by applying methodological rigour in empirical studies abstract truths about reality can be extracted and distilled, bring a closer approximation to ‘reality’ (Law and Urry 2004:13). This builds on an extrapolation of Mode 1, that there exists one external reality. Mode 2 science studies does not deny the possibility of one external reality, rather it questions to what extend consensus can be achieved at all. As such, Mode 2 deconstructed science in such a way that it showed the social constructed nature of science; making all statements about reality relative. Mode 3 science studies, which sometimes is called the material turn in science studies (see Bennett and Joyce 2013), has softened this proposition somewhat. In that it reintroduced realities influence (i.e. materiality). However, it still reserves the determining power to the social constructed nature. This means that ‘reality’ is the product of the ‘real world’ plus social constructions (c.f. Latour 1999). This applies to all human endeavours, including all branches of science (and the social sciences).

With this in mind, the way consolidation is reached is through social consensus (hence social construction being attributed this determining power, c.f. Harraway 1991, Latour and Woolgar 1979; Shapin, Schaffer et al. 1985; Yearley 2004). Different mechanisms have been described how this process is achieved in practice. One way this can be achieved, that I want to highlight here, can be conceptualized by using an ‘economic metaphor’. This is a redefinition of the Mode 1’s canon; ‘nanos gigantum humeris insidentes.’[5] In this view the giants are not only aids, but also represent powerful allies. The cost that is involved in unmaking the newly created network (argument + alliances) determines if a proposition is accepted as true or false, criticised or praised (Latour and Woolgar 2013:243). The more alliances that can be mobilized, the harder it becomes to defeat the proposition. Once enough allies have become (rhetorically) mobilized a statement is accepted as true, i.e. it is logical (Latour 1987),[6] this applies to both the NSS.[7]

The third and last, similarity I want to touch upon between the sciences and the social sciences is that of the framework in which they operate in. This is probably the aspect in which the social and natural sciences are most similar to each other. Because, regardless of what research subject you look at, most research today; is done at universities, has to apply for research grants, has to be approved by ethics committees, and has to go through the peer-review process to just name a few. According to Mode 2, these structural processes are not neutral. Examples of the influences of the aspects that can be mentioned are: the masculinisation of the concept used in science (Martin 1991), the effects of strutural changes (Abbott in Camic, Gross et al. 2012:43-88) or the Euro-American domination of research (Law 2004). As said, this is most likley true for the both natural and social sciences. However, how this process is influencing the social sciences in particular and more importantly, what influences do we want this to have is still a big question mark according to Law and Urry 2004.

References


               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               
               



[1] ‘Rediscovering, what humans have always known, this is what we call science’  translation by the author
[2] Mode 1: is regarded as the philosophy of science, which was ended with the publication of Kuhn’s (1970) The Structure of Scientific Revolutions
Mode 2: was primarily occupied with socially deconstructing (hard) science, examples are; Latour and Woolgar 1979; Bloor 1991 or Collins 1981
Mode 3: emerged with the arrival of STS and the publication of Leviathan and the Air-pump (Shapin and Schaffer 1985), but was consolidated after the so called Science Wars (c.f. Gieryn 1999) with the introduction of the so called material turn. This approach (and other) tries to remarry social construction with expertize (C.f. Collins 2010) and the ‘real’ (material) world (c.f. Latour 2013)
For a more detailed definition of the different modes of science studies, see Collins and Evans (2002)
[3] The German word ‘Wissenschaft’ cannot be directly translated to the English word ‘Science’ as it applies to any study or science that involves systematic research, including social sciences and the humanities.
[4] The term physics envy is used to criticize a tendency (perceived or real) of softer sciences and liberal arts to try to obtain mathematical expressions of their fundamental concepts, as an attempt to move them closer to harder sciences, particularly physics and move up on the ‘certainty scale’.
[5] English translation: dwarfs standing on the shoulders of giants
[6] Hence, there is no real tension between facts being ‘invented’ and facts being ‘true’
[7] A difference between the two is that, social science claims are less costly to unmake than science ones. For example convincing someone that the billion Euros spend in building the Large Hadron Collider is a futile enterprise is harder than convincing someone that a qualitative interview study is, as more allies have been mobilized. This is another contributing factor which creates this apparent hierarchy between the sciences and social sciences. 

24/01/2015

Application of the Radical Programme to the Hygiene Hypothesis



26-05-2011

Application of the Radical Programme
to the Hygiene Hypothesis
by René Brauer

Gothenburg University
Institutionen för Filosofi, lingvistik och vetenskapsteori
VT2107 Modern vetenskapsteori och vetenskapsstudier
Supervisor: Henrik Lundberg

In 1989, David Strachan formulated the Hygiene Hypothesis (HH) after observing a correlation between family size and the prevalence of Hay Fever. The more siblings a child had the less prone it was to develop Hay Fever. He hypothesized that contact with the pathogens in a early age can built resistance for the developing child preventing the outbreak of allergies. (Mutis 2007, p. 433; hereafter A + page number) Allergies in general have been on the rise in the developed countries in the past decades. While viral and bacterial infections have been declining due to better hygiene standards and advanced medication. The HH links these two phenomena together and states that because of the rise of more hygienic environment and advanced medication there has been an increase in allergic diseases, because the immune system cannot develop sufficient resistance in early life. (Garn and Renz 2007, p. 442; hereafter B + page number)

The assumption of HH has developed into a working hypothesis of several epidemiologists, since it promises a viable explanation to the increase in allergic diseases. It holds the promise to develop possible mechanism of prevention and medication (B, p.448) and therefore can attract real public interest and possible sponsoring. Making it a proxy for how science is done today. That the scientist is working in an environment that is dependent not only on factual issues but also on social funding. (Shrader-Frechette 1994, pp. 98–100) The HH is a relative young working hypothesis, so entrenched polarization has not yet occurred in the subfield. The issues of HH are still on a very fundamental level, being disputes over how the underlying mechanisms work. (A, p.434) Nevertheless underlying doubts still can be raised. This seems reminiscent of the development of the revolutionary progress of science in the spirit of Thomas Kuhn. (Chalmers 1999, p. 108)  So the question poses how is this development progressing in HH?

Working Questions

·         How has the Hygiene Hypothesis developed since it was first proposed in 1989?

Methodology


The conceptual framework used in this study is that of the Radical Programme (RP) introduced by Harry Collins (1981). This allows for a working framework without going into depth into the subject studied. Since it is not concerned with establishing which faction of the argument is in the ‘right

A big weakness of RP is shared with the Strong Programme (SP) of David Bloor (1991) that it has an under-determination into its social dimension of its explanation power, relying on the interpretation of the researcher. (Woolgar 1981, p 381) The Duhem-Quine Thesis states that a theory can never be sufficiently determined by their data, the same is true for the explanation giving by RP and SP of its social dimension, since there could be a myriad of different social reasons leading to the same action. However since this study is neglecting the social explanation side and only wants to look at the development of a scientific working hypothesis, this is something that possible continued research has to address. In this study it is of minor consequence and in line with Collins argumentation how any work in the sociology of science can be done at all. (Collins 1981, pp. 216-217)

The study has been conducted as a literature review of two articles (A and B), which in themselves are summarizations on the field of epidemiology, regarding the findings of the HH. The articles have been analyzed in the spirit of content analysis. Summarizing the articles in discrete categorizes allowing for comparison and comprehension of the text material. (Weber 1985) The summarization has been structured according to the TRASP, principle in RP, meaning there will be a neglect to establish which opinion is ‘right’. (Yearley 2005, pp. 28 - 29) Since the aim of the study was not to proof if the stated beliefs are true or not, but merely to see how the HH has developed. On top of that there is no way for the researching sociologist to be able to make this kind of judgment anyway (Collins 1981, p. 216)

Presentation of the Articles


Since both articles are in themselves literature reviews, I have chosen to present the underlying picture of both articles rather than each article separately. The current definition of the Hygiene Hypothesis has been aptly put as: 

that certain environmental factors, to which children are exposed in early–life, interact with the specific genotype of a child and influence the developing immune system in a way that either predisposes to or protects from the development of an allergic disease
(Hoffjan et al., 2005 in B,p. 442)

The increase of allergies is a well established fact, (A, p. 433) several studies have shown it and given credence to the HH. Studies have showing the relationship between daycare attendance and allergy frequency is beneficial. The underlying principle has been proposed to be the same as of the HH, the unhygienic contact with other children can boost immune systems defense. (B, p. 433) Children living in rural areas on farms have been compared with children in the villages that have less contact with the farms and served as natural controls. The same correlation in regard to a low frequency of allergies was found. (B, p. 445) A Swedish study showed that pet ownership decreased the outbreak of allergies in children (B, p.443), even dietary choice which prefer food with high bacteria content like unpasteurized milk and fermented vegetables have proven to show protective capabilities in regard to allergies. (B, p.442) Lower antibiotics consumption is also in correlation with lower frequency in allergies, (B, p.442) even if these findings are still disputable. Both are thought to have an effect because the gut flora is important for the development of the immune system.  One of the most striking examples for the HH is the prevalence of allergies in Former East-Germany. Until the end of the Second World War there was no significant difference nationwide in allergies compared to the western part. Studies have shown that in 1990 allergy frequencies where higher in West-Germany than in East-Germany, only to normalize to the higher level in all of Germany today. This is explained with East-Germany having lower economy standard and respectively lower hygiene conditions. (B, p.443) This wealth of data makes a very strong circumstantial case for the HH.

Several explanations have been proposed in understanding how allergies work. The overall consensus is that genetics alone is insufficient in explaining the outbreak of allergies in these different quantities. In other words allergies develop because of the interaction between environmental factors and genetic factors. (B, p. 442) This transformed the original straight forward HH into a far more complex idea, been placed in the middle of the nature/nurture debate. (A, p. 435) The underlying mechanism instead of being a simple causal chain are one of many interlocking causal correlation that have to work at the right time and quantity with a myriad of external factors satisfied, to be beneficial.

This increased complexity opens up for flexibility in the different interpretation of the HH and even entire different interpretations. One worth to mention is the ‘pollutant theory’ where instead of more hygienic standards, new exotic pollutants created by our modern lifestyle are responsible for the increase in allergies. (B, pp. 442, 447)

The environments studied are very complex, for example in the Germany case that deals with a entire nation consisting of several million individuals. It is almost impossible to deduce all possible variables regarding such a large population size. Having said this, the HH still seems as a viable explanation, since several underlying microbiological mechanism have been found to explain the assumption. (B, pp. 445-446) The HH is still in its early development that no practical implications can be drawn, despite what the lay press may say. (A, p. 437)

 

Analysis

The analysis is presented in the different points in how the HH has developed; note these are not in chronological order. 


  • Finding that several genes are responsible for the outbreak of an allergy combined with environment interaction makes it even more complex. (A, p. 434) This in term lead to the discovery that there are different phenotypes within the HH, meaning that individual A could have a beneficial effect while individual B would not, so to speak an individual had to be primed to be perceptible for the effect described in the HH. (A, p. 436)
  • As well as the exposure being dosage dependent, too little exposure had no effect contrasted to too exposure that could lead to an outbreak of an allergy. (B, p. 446)
  • Even some still argue that allergy once manifested is less severe than it otherwise would have been. (A, p. 435). Furthermore the same exposure can have different effect on different phenotypes (A, p. 436). 
  • The receptors which bind to these in the cell have been show to be multiple different, making the relationship between them important too. (B, p. 447)
  • The emergence of a timing window is steadily manifesting. Meaning that the earlier the exposure was the greater was the benefit, some even reporting ‘exposure’ in utero as beneficial (B, p. 442). Furthermore different allergies having different timing windows. (B, p. 447)
  • Not only direct exposure to the pathological, bacteria, components or direct causes of the allergies had an effect. Even other substances showed similar effects. This is something that was not included in the initial HH. (B, p. 446)


These were some of the developments that have been made in the HH that partly were not hypothesized by the original idea, but subsequently have been shown to have an effect.There is some methodological consideration to be raised on the studies presented in one of the article. One study for example focused on atopic asthma in children (A, p. 434). A big problem in this kind of study is that the reporting was done by parents and wheezing was used as an indicator. However wheezing is caused by several different types of asthma not all of them being allergies. (A, pp. 434-435) Furthermore the decisions were made by parents and not medically trained personal this leaves for an inherent insecurity in the data. It is conceivable that the other studies if presented in more detail would show similar problems. 


Discussion

As seen above the HH has steadily increased in complexity ever since it was first proposed in 1989. Today the reasoning is not only including pollutants that cause allergies, but also viruses, bacteria being of pathogenic character or not. This shows a clear deviation from the initial position. Showing that once a working hypothesis has been established, more data will ‘fit’ the explanation. Otherwise these observation where just random, giving credence to Collins proposal that data outside an interpretative context is meaningless. (Collins 1983, p. 91) Once a working hypothesis is established, HH in this case, the study can commence. Much like the notion of puzzles in the Kuhnian view of paradigms. There needs to be a framework on what data can be interpreted to be able to make any progress at all. (Chalmers 1999, p. 110)

As mentioned in the summary of the articles, there is not of yet a clear polarization in this particular field between two or more different theories. For now this is confined to a dispute or discussion of the working or the underlying principles. But as we have seen there is a lot of uncertainty and flexibility in the interpretation that could give leeway to these different interpretation and possible formation of a controversy. Since all theories are always simplifications of reality and reality being far too complex to be able to determine, some of these uncertainties will never disappear.   


References 

Article A: von Mutis, Erika (2007): Allergies, Infections and the hygiene hypothesis – The epidemiological evidence, , Immunobiology I. 212, pp 433-439

Article B: Gran, Holger & Renz, Harald (2007): Epidemiological and immunological evidence for the hygiene hypothesis, Immunobiology I. 212, pp. 441- 452

Bloor, David (1991): Knowledge and Social Imagery, Chicago, The University Press Chicago

Chalmers, A.F. (1999): What is this thing called Science? 3rd edition. Indianapolis/ Cambridge: Hackett Publishing Company, Inc.

Collins, Harry (1981): What is TRASP? The Radical Programme as a Methodological Imperative, Philosophy of the Social Science, 11, pp. 215 – 224

Collins, Harry (1983): An Empirical Relativist Programme in the Sociology of Scientific Knowledge, in Knorr-Cetina, Karin and Mulkay, Michael (red) Science Observed, pp. 85-113

Hoffjan, S., Nicolae, D., Ostrovnaya, I., Roberg, K., Evans, M., Mirel, D.B., Steiner, L., Walker, K., Shult, P., Gangnon, R.E., Gern, J.E., Martinez, F.D., Lemanske, R.F., Ober, C., (2005): Gene-environment interaction effects on the development of immune responses in the 1st year of life. American Journal of Human Genetics 76, 696–704.

Shrader-Frechette, K., (1994): Handling Conflicts Through Stage-Two Ethical Analysis: Giving Priority to the Common Good. In Ethics of Scientific Research. Lanham, Md: Rowman & Littlefield. Ch. 5.

Weber, Robert Philip (1985): Basic Content Analysis, Beverly Hills, SAGE Publications Inc.

Yearley, Steven (2005): Making Sense of Science: Understanding of the Social Study of Science, London, SAGE Publications Inc.