Monday, August 9, 2010

Science - Subjective or Objective

A Response to Helen Longino's Paper "Values and Objectivity"
 
The objectivity of science and of scientific inquiry is often perceived as vital to its ability to uncover truth. This objectivity is seen as a way to ensure that the explanation of the nature of an object is independent of any features of the particular subject who studied it. This means that an objective account of the nature of an object does not draw on any assumptions, values or prejudices of a particular subject and thus can be universally accepted.
Philosophers such as Thomas Kuhn[1] believe that science can never truly be objective, saying instead that scientific knowledge is simply knowledge of the nature of objects as we perceive them, not as they actually are. They argue that perfect objectivity can never be achieved in science, as all scientific methods of inquiry are based on human concepts, tools and thought patterns. This direction of thinking tends towards issues revolving around scientific realism; the concept that science describes truths about the real world.
For the sake of this paper I am not going to concern myself with issues of scientific realism. Instead I will take on a varied form of a constructive empiricist stance, stating that we are not concerned with absolute truths within the real world. I will look at the objectivity of science from the stance that we are interested in determining the truth of the nature of objects as we perceive them. Future use of the word truth throughout the paper will be in reference to this concept; that we are seeking truth relative to the way we perceive the world.
I believe that it is because of the ambiguous nature of the definition of objectivity that the objectivity of science has come under attack by philosophers such as Thomas Kuhn1 and Paul Feyerabend[2].
If we define objectivity as a way to seek truths relative to the way humans perceive the world, we are accepting that objectivity is a human construct in itself. This form of objectivity is a cornerstone of science. If scientists allowed their own personal bias to influence their work, their findings would be misrepresentations of the actual truth and thus detrimental to attainment of knowledge of the nature of objects as we perceive them. Objectivity in science will therefore increase the reliability of scientific knowledge, but it will not grant science infallibility or any form of knowledge on the nature of objects from all perspectives.
In this paper I will look in particular at Helen Longino’s account of objectivity in science[3], specifically addressing her claim that the objectivity of science is secure because of the social structures surrounding modern scientific inquiry.
An important distinction to make of science is that it is not simply a body of knowledge, but instead the practice and methodology of how that knowledge is gained. So when discussing the subjective or objective nature of science, we are really discussing whether the practise of science, or more correctly the scientific method, is objective or not. We are to presume that if we were able to form a completely objective scientific method, then the knowledge gained from such methodology would be equally objective, as the knowledge would have been independent of any personal bias.
Longino suggests that by defining science as a practice we can now show that it is not something that is practised by an individual but instead by social groups. This social nature of science is emphasized by Marjorie Grene[4], where she speaks of scientists’ reliance on others for their education, ideas and instruments. With the rapid expansion in the wealth of our knowledge it is almost a necessity to have several different scientists with varying expertise working together on a project. A look at a project group developing a vaccine for malaria demonstrates this, with their team being comprised of medical professors, structural biologists, parasitologists, and members from the Department of Biochemistry and Department of Molecular Engineering of Monash University, as well as the engineers required to operate the complicated machinery such as the Australian Synchrotron[5]. Longino calls this “‘big science’ ;(where) a single complex experiment may be broken into parts, each of which will be charged to a different individual or group of individuals”.
Longino argues that while it may still be shown that the individuals are in at least principle practising science, scientific knowledge is gained not just by collecting all the data into one whole, but by the critical evaluation and modification of that data by the scientific community. This “critical emendation and modification”[6] of an individual’s data by the society is what Longino believes ensures that science is still objective. Processes such as peer-review analyse papers to ensure that they are methodologically correct and that the data is sufficient to support the conclusions made. Further repetition of experiments once the papers are published ensure that the same results can be found in varying labs by individuals whom hold differing values to the initial investigators. Even further down the track, when hypotheses based on knowledge gained from the initial findings don’t hold true, then those original findings are often scrutinised further. This gives science its perception of being self correcting.
The idea that in science nothing is sacred demonstrates that even for a well accepted theory, given enough criticism or the presentation of another theory with even greater supporting evidence, the scientific community will amend their concepts on the nature of the relevant objects. Examples of this are scattered throughout history, with Newton’s universal law of gravity standing strong for several hundred years until varying scientists began to criticise and eventually modify the theory to include Einstein’s theory of general relativity.
Do such processes really provide science with its objectivity? There are obvious benefits to such a system; it almost guarantees that any methodological errors will be picked up by the many checking stations along the way. It also takes the pressure of the individual scientists to be completely unbiased, an unrealistic expectation considering the complexity of the human mind and our varying social and cultural prejudices.
There now may seem to be no need to be unbiased in their methodology either, as any such errors would be resultantly picked up by peer-review or any of the other many checking stations. This, however, would lead to a seemingly very inefficient form of science, in which scientists could present sloppy work, including calculation errors, or not provide substantial data to support their conclusions. These errors could take substantial time to detect and would thus greatly slow the attainment of scientific knowledge. This of course is not a required artefact of accepting Longino’s account of the social structures protecting objective science and can be easily avoided by still promoting a level of quality in a scientist’s work, ensuring that studies are “double-blinded” and that there is always a complete disclosure of results.
The benefit would instead be that the scientists are free to explore avenues of research that they value, as is already commonly done today in the scientific community with issues such as climate change and breast cancer research, without the fear that their personal values will affect the truth value of their research. It would also promote a form of creativity among the science community, as a scientist with vested interest in his or her hypothesis is likely to actively defend it against criticism.
As stated above, these types of actions already seem prevalent in the scientific community and are most likely inescapable when dealing with a species such as ours, due to our tendency towards subjectivity. So Longino’s account of the objectivity of science, in particular the scientific method, seems to allow for an individual’s idiosyncrasies yet still permits science to be objective if it is practised by the community.
The problem with such an account is that Longino is vague on outlining what level of criticism is needed before a hypothesis is recognised as objective. Longino states that “The greater the number of different points of view included in a community, the more likely its scientific practices will be objective, that is, that it will result in descriptions and explanations of natural processes that are more reliable in the sense of less characterised by idiosyncratic subjective preferences of community members”[7]. You could look at her account in two ways.
Firstly you could understand Longino as saying that objectivity requires a constant level criticism and infinite view points. Anything less, while it may have a degree of objectivity, is not truly objective. If this is the case then her account differs very little from those who perceive true objectivity as an impossibility for science, as any definition that requires an infinite and continual level of criticism from an infinite source of views is unrealistic and the definition of no further use.
If on the other hand, she believes that science can be truly objective after a certain degree of criticism, but then offers no account for what level of criticism this equates to, or why such a level provides objectivity. One might respond to such criticism by saying that Longino was not referring to single experiments or research projects being objective in nature, but simply that as a whole. The scientific method is objective if it continues to critically evaluate and modify existing theories and hypotheses. If this were to be the case then she would have had created a paradox with her original statements regarding the objectivity of scientific knowledge being due to the objectivity of the scientific method. Longino would be left with an account of an objective scientific method that cannot produce objective knowledge without an infinite timeframe, and so our current knowledge of the nature of objects is no more objective under Longino’s definition then it would have been under any other.
Another issue with Longino’s account of objectivity is actually provided by her, as she describes that the belief we can just read data is a recipe for replicating mainstream values and ideology[8]. As shown, Longino accepts that individuals cannot escape the confines of subjectivity and are inevitability going to impose their personal bias on events. But what is it that makes a community so immune to such a bias? Why is the chosen criticism of someone’s hypothesis not biased in itself. She claims that the combined critical review of others’ work by a community will negate any such bias, but in doing so she is stating that objectivity is decided by agreement rather than actual truth. Knowledge is unquestionably benefited by the contributions of a society, but a critical consensus should not be the standard for objectivity. Objectivity cannot be measured by the cognitive needs of a society and must be defined by what actually is.
Processes such as peer-review and the many other checking stations of science do a play an important role in the scientific method and can in some instances remove bias, but that is just an artefact of the error checking process that peer-review is. A scientific paper should not need to conform to any particular values to be published, it should just be error free and ultimately that is all peer-review should be able to analyse. The current state of the peer-review process actually reflects on the problems presented by Longino’s account of objectivity, with a raise in concern for the criteria needed for a paper to be published in several of the mainstream journals. Longino talks about the breakdown of the peer-review process but describes peer-review as a way to “bring to bear another point of view on the phenomenon”[9]. She describes the situation where a reviewer assumes that a person would not have a job at X institution if they were not a good scientist and thus their work could be presumed to be of good quality. This situation she claims, while not desirable, is alright because further scrutiny after publication would compensate for any failings of the peer-review process. What she fails to address are not situations where papers are published because of values the reviewers hold, but instead where papers are discredited because they don’t align to the reviewer’s values. Objectivity should not be defined in such a way that it relies on the belief of the community. “It is inadequate because, believing, like wishing, does not make things so”[10].
A possible solution to the issue concerning the definition of objectivity is to show that objective knowledge, achieved through objective methodology, does not have to necessarily equate to truth. One could objectively analyse the provided data and come to a certain conclusion that when more data is provided can be seen to be wrong, but this does not prevent the original analyses from being objective. An objective explanation would therefore entail that it was concluded using the most reliable methods available at the current time. A definition of this form would enable science to strive to be objective by simply being reliable.
Conclusion
Objectivity in science, particularly in the scientific method, has long been held as being a founding principle of what science is, but difficulties in defining objectivity has made it difficult to outline how to attain a truly objective scientific method. Past attempts to provide objective conceptual methodology by some philosophers have resulted in paradoxical definitions of objectivity, whilst other philosophers have defined objectivity in such a way that it is unattainable in science, a definition that ultimately is of no practical use.
Helen Longino’s defence of the objectivity of science is flawed in several ways. Her account of objectivity in scientific methodology is either paradoxical or of no use due to its unrealistic requirements, depending on how you interpret what she says. Longino suggestion that the standard for objectivity lies in the views of other people is flawed as it provides no explanation for why the criticism of biased views cannot also be biased in itself. A society’s belief in the true nature of objects should not be taken as an objective explanation of the true nature of an object.
I believe the answer lies in an appreciation of the fact that as humans we are limited to understanding the nature of objects only as we perceive them and that objectivity itself is still a human construct. Therefore, it is only of use if we are to define it in such a way that shows it is for seeking the truth of the nature of objects, relative to how we perceive them.
We must also not limit objective knowledge (knowledge that is gained through an objective analysis) to being equal to the truth of how objects are. This would require a form of omniscience that is unrealistic to ask. Instead we must see that an objective account is one that reliably defines the nature of an object to the best of our current understanding.

No comments:

Post a Comment