Question Submission 12

Please add your question as a comment to this post.


Using the guidelines set forth by Schaffner, Hull lays out numerous examples throughout Chapter 1 of his 1974 book to demonstrate the difficulty in reducing Mendelian genetics to molecular genetics. Although certain examples meet Schaffner’s specifications, there are many that fail to do so (e.g. the dominant/incomplete dominance scenario described at the top of pg. 41). Hull summarizes these difficulties in the following statement: “Phenomena characterized by a single Mendelian predicate term can be produced by several different types of molecular mechanisms. Hence, any possible reduction will be complex. Conversely, the same types of molecular mechanism can produce phenomena that must be characterized by different Mendelian predicate terms. Hence, reduction is impossible (pg. 39).” Hull then states that the latter examples could be straightened out, but doing so would be more an act of replacement than reduction.

Thus, reducing Mendelian genetics to molecular genetics, if possible, would no doubt be an arduous task. However, my lasting impression of this passage dealt with the implications/consequences of such a reduction. Hull states on pg. 44 that: “the effort to reduce Mendelian genetics to molecular genetics hardly seems worth the effort. No one is currently engaged in producing such a derivation, and there seems no good reason to do so.” What, then, is the implication of either: (a) the inability to make the reduction, or (b) successfully making the reduction? Certainly, one could argue that both Mendelian genetics and molecular genetics have been, in their own rights, successful. If any such reduction would only change Mendelian genetics in a way that was analogous to the current system (as specified by Schaffner), what would really change between how the two systems are currently viewed? Is there truly, as Hull states, “no good reason” to make such a reduction?

Toward the beginning of Chapter 1, Hull claims that after forming rigorous definitions of "reductionism", philosophers are unable to classify any scientific movement as solely undergoing reduction. However, he also says that many scientific revolutions have taken place. He does not mention the definition of these revolutions. What type of definition can we give for a scientific revolution? Could it be possible that in developing a rigorous definition of "scientific revolution" we may also not be able to classify any scientific movements under this title?

I am also curious as to the relation of these two concepts. It seems as though Hull is comparing them; however, it feels as though the reductions may be a subset of scientific revolutions.

In Schaffner's article, "The Watson-Crick Model and Reductionism," he claims that in order to have successful reduction functions, "one needs experiments that show that the entities, independently characterisable in terms of the reduced and reducing theories, and which are to be identified, have the same or relatable relevant properties." This leads to a question that I have had while thinking about reductionism. Is experimental work ever done with the predetermined purpose of reduction? I usually think of reduction as something done unconsciously by scientists, that is later brought to light by philosophers or historians. Is it common for scientists to consciously set out to make reductions?

As I read through Hull's first chapter I couldn't help wondering where scientific practice fits into the replacement/reduction equation. Hull refers to the issue as 'theory reduction', but I wasn't convinced that this was so much an attempt to purposefully background issues of practice as it was a case of what Waters would call a theory-centric bias. All of the angles that Hull took on what constitutes a replacement or a reduction were based on the theoretical framework that each field was functioning in, i.e the possibility for analogous theoretical terms, idealistic simplicity and a wide breadth of explanatory power, etc. So, technically, if these theoretical requirements were met one field could be reduced to another whose practice does not resemble the reduced theory's practice at all. What does this say about the vastly different practices that were employed between Mendelian or classical genetics and molecular genetics? What could we say about the possibility of reduction if we take into account this disparity? What would a counter argument from the perspective of a practice-based epistemology such as the one formulated by Waters be?

Hull's analysis of the reduction of Mendelian Genetics to molecular genetics ignores what I think may be a central issue in discussions of reduction in biology. Essentially, Hull attempts to describe the reduction of one entire domain, modern transmission genetics, to another, modern molecular genetics. However, according to the reduction criteria provided by Nagel, Woodger, Quine and Shaffner as outlined on pages 31 - 33 of Hull's book, reduction is predominately between two theories within a domain. Why does Hull require that the theories involved in the reduction be the grand central theories attributed to the domain? Isn't it consistent with the reductive program to say that some theory or explanation in transmission genetics can be reduced to another in molecular genetics, but it may not be possible to reduce the central theory of each domain (if there is such a classifiable thing) nor can it be done for the entire domain, whatever it may mean to reduce a domain? Where did this attempt to reduce the entire domain come from, even if post-hoc corrections to the domain are made so as to be favorable to the reduction?

The attempt to reduce entire domains appears incoherent due to the complexity of scientific domains, the overlap between related domains, and the overlap of techniques, instrumentation, and investigative strategy. Not to mention that issues of non-arbitrarily, yet precisely characterizing the specifics of scientific domains and their objects of investigation seem to plague this account from the outset (Hull certainly does an excellent job of making that clear, but I can't seem to get passed it).

Hull explains Kuhn’s definitions of the terms ‘revolutionary science’ and ‘normal science’ attributing sweeping, replacement changes to the former and more plodding, minor changes to the latter term. Hull offers the example of Aristotelian physics ultimately being replaced by Newtonian physics as a clear example of revolutionary science. But to me, this concept of a science outright replacing another is misleading; I believe that ‘refining’ is a better description. To speak of revolutionary science in terms of replacement seems to suggest that an old science being replaced is irrelevant while it is in fact a critical stepping stone in reaching the newer, more refined, and accurate depiction of science and is worth our attention. Moreover, to understand the state of current science, it is critical to understand how science has evolved and what the old, unrefined, perspectives of science were. This is especially important to consider when we appreciate (assume?) the fact that the current theories of science are incomplete and riddled with inaccuracies and incompleteness, and by understanding the incorrigible thought processes that birthed modern theory, science stands a better chance of mending our scientific fallacies.
Does it not seem that describing revolutionary science in terms of replacement, and not refinement, neglects the importance and relevance of the progress that has come before? Or am I seeing a problem where one does not actually exist for thinking of revolutionary science in terms of replacement actually does not disregard history?

My question for this week is a bit of an aside, having to do with a philosophical position that Hull brought up in this first chapter. He mentions something called an "operational definition", or an "operational position" and I am curious as to what sort of role this line of thought played in any of the philosophy of science literature. From what the Stanford Encyclopedia of Philosophy had to say on "Operationalism", it is a position that is considered "extreme" and "outdated" and I would like to explore why. It seems as though when we are talking about something like electrons, the definition of the term is very much dependent on the context in which the concept is being discussed. In one theory what we mean is a negatively charged particle, in others a point with no mass, and in still others a smear. While I can understand this being an 'extreme' case, I wonder if there is not some substantial connection between the idea of a practice-driven science and some basic form of this operational definition of terms? However, there is the distinct possibility that I seriously misunderstand some consequential baggage coming in through the door with operationalism.

So after reading the Hull article I am very confused about what it means to be a reductionist and why anyone would want (or not want) to be one. I think I understand that it is a question about the relationship between different parts of knowledge that might also reflect the mereology of nature, but I realize now I don't understand the history of the philosophical debate. Hull claims in the conclusion to have explained why the positivists cared about reductionism, but I am pretty sure that he did not. Both the question of inter-level (mereology) and inter-theoretic (knowledge through time) seem to be important to the question of the rationality of science.

What I am wondering is how being a reductionist or not of any variety (ontological, epistemological, methodological) manifests itself in action in addition to thought?

Bill asked some of us last week what it means to be a methodological reductionist and I think this question was what he was getting at. I had no idea then and I still do not. I think that understanding the history of the philosophical issue would make this salient.

In Ch. 1 of Hull's paper I appreciate the idea that science is a messy subject at times and doesn't have all the answers, yet can provide plenty of logical answers for many problems. The question I have comes from the idea that a theory is an unobservable "entity and process". Hull goes on to say that there are however, theories that can be observed in some ways, but these theories are no more theories than unobserved theories. What then makes a theory truly strong: what lies beyond an observable theory? A fact?

To what extent is the difficulty caused by the reduction of Mendelian genetics to molecular genetics caused by the fact that neither are formal theories, like those of physics? In both selections on reduction in biology so far, a physical reduction has been given as a paradigm and then compared to biology. What examples of biological reduction do we have? Or examples of non-formal/non-mathematical reduction? Is it possible that if biological reduction is possible that a separate model of reduction is necessary? If the sciences in question (biology as compared to physics) do not function in similar ways, then it seems like a rash assumption to think that reduction between theories would function similarly in them. I would like to hear arguments for and against the lack of distinction between the sciences where reduction is concerned.

Hull brings up the idea of 'revolutionary science' and 'normal science.' I don't really see a need for the distinction between 'revolutionary' and 'normal' science. In both cases, changes are happening in the sciences. A theory is either getting completely replaced or changed just a little bit. Even Hull admits that, "Neither the distinction between revolutionary and normal science nor the distinction between replacement and reduction is sharp." Why is there an insistence on making a distinction between 'revolutionary science' and 'normal science'? Whether a science has a huge impact or a small impact, the science still made a change and has made an impact, regardless of how insignificant the change may be. It just seems to me this is a way for scientists to say that their work is important by calling it 'revolutionary science' rather than 'normal science'.

We started to touch on this subject in last Tuesdays class, but I am having trouble making the connections between reductionism and its importance/role it plays in modern science. It is stated in Hull's paper that "All changes in science are really replacements. Reduction in any strict sense is impossible" (p11) and yet the paper goes on to discuss reduction theories in more detail. After "reduction" has been defined, what do we gain from it? Where does this analysis become beneficial?

In “Reduction of Mendelian Genetics” Hull discusses the history of scientific reduction. He says that in the past all reductions have occurred within a single, well-defined area of science. Only in the past two decades has reduction occurred across multiple disciplines. The example he mentions is the reduction in biology of Mendelian genetics to molecular genetics. My question is about his statement that prior to the last two decades no scientific reduction occurred across different areas of science. To me, this seems nearly impossible. How could one explain theories in biology without using some chemistry or other areas of science? Even if you are not discussing chemistry concepts in a biology theory, the chemical elements are still there and still very much a part of the theory.

On page 38 Hull stats, "If Mendelian genetics is to be reduced to molecular genetics, then better reduction functions will have to be formulated. However, the formulation of such reduction functions turns out to be a good deal more formidable than one might at first expect." One of the predominant obstacles confronting this reduction is the differentiation of terminology between fields. This differentiation appears to have been facilitated by the separation of heredity and development, and as a result, seems to me to be a persistent source of failure in reducing Mendelian genetics to molecular genetics. The reason for its persistence sprouts from the distinct difference of objective between Mendelian and molecular genetics, understanding heredity patterns and formalizing developmental processes. Is it not? Doesn't the terminological gap between the fields sprout from their distinct objectives? Hull claims that to reconcile the terminology would likely involve restructuring much of one field or the other, or both. Firstly, is that a realistic venture? Secondly, would there be any worthwhile result from such a venture? If so, what might that be?

Hull finishes the chapter with a provocatively simple claim, "all one has to do is to provide the reduction functions necessary for the derivation of Mendelian genetics from molecular genetics, carry out the derivation, and then present an analysis of "strong analogy" such that the corrected version of Mendelian genetics is strongly analogous to the uncorrected version." His reversal in tone from skeptic to disciple shocked me a little. It got me thinking...rather then reducing Mendelian genetics to molecular genetics, could molecular genetics be reduced to Mendelian genetics? Can developmental processes be deduced from hereditary patterns?

This week I have two questions. The first is related to the statement about Mendelian genetics, a "biological theory" being replaced by molecular genetics, something they state is a more physico/chemical theory. Personally as a biologist, I feel that this statement is false after taking genetics. I would say that it is still ruled by biology, and can't it also be argued that physics and chemistry have been integrated into biology? Physics uses calculus yet we call it physics. Can biology not use physics and chemistry in the same sense without being called out for being not biological?

The second question regards the distinction between revolutionary and normal science. I personally think that all scientific discoveries are revolutionary because they all essentially have a profound effect on the way science is perceived and passed on. Since there is no fine line between what Kuhn would call normal and revolutionary science, can't we just eliminate the distinction all together?

My question this week has to do with the distinction that Hull makes in his Chapter One between the historical view of reduction and the rational view of reduction. Hull says the former is concerned with the historical process of scientific theories as they were formulated. The latter, according to Hull, is viewed formally as an investigation into the relations which emerge between two scientific theories, and then reconstructed after being extracted from their historical settings.
So, I am confused... with:
Why someone would choose one view over the other view? (as defined by Hull)
But more specifically, what would be the end consequences to the acquisition of one's knowledge upon one's acceptance of the historical reduction view or the rational reduction view?

It seems to me to an extend Hull is trying to differentiate between Mendelian genetics and Molecular genetics - because Mendelian genetics assumed that genes were stable, that position did not matter, that genes were either dominant or recessive and had not proper way to account for epistasis and pleiotropy. Ok. Then, one could say that Mendelian genetics was a form of naive molecular genetics -and as some of these complexities were recognised scientists moved to molecular genetics. Then again Hull seems to suggest that adherence to one theory or another are not entirely a result of available evidence. How else do we explain the move from mendelian genetics to molecular genetics - when we can see mendelian genetics as a subset of molecular genetics or as an indealised case of molecular genetics?

After reading Hull's "Reduction of Mendelian Genetics," I find myself wondering why reduction is necessary and wondering if it is really possible. I just don't understand why it is necessary to reduce a theory? It seems like more of a hassle than anything. I find a replacement theory to be more useful than trying to reduce them. I think that it would be easiest and most useful to leave topics as independent of each other.
Hull makes a comment about there is no clear point where Mendelian genetics ends and molecular genetics begins. If this is one of the criteria to make a theory reducible, is it ever really clear where one topic end and the other begins? I would assume to there would have to be some exceptions in reducing theories, so why create the confusion?

This is an excellent perspective, and I believe it summarizes many of the frustrations/confusion we have been discussing during this semester. Why is reduction necessary? And if one does reduce a theory, what is the correct way in which to do so? I personally feel that reduction can be taken too far. The disadvantage of taking reduction too far is losing the knowledge that can be discovered during the exploratory process. Also, too much reduction seems to shift the focus too heavily on defining theories rather than new discovering knowledge.

"Ok. Then, one could say that Mendelian genetics was a form of naive molecular genetics -and as some of these complexities were recognised scientists moved to molecular genetics."

I think some of these complexities were recognized by earlier scientists but these issues could not be examined due to limited technology of the time. The rise of Molecular Genetics was, in my view, the examination of the same systems with more advanced apparatus and a more developed way of manipulating the system(s).

In regards to your question I think it gets down to what is at the core of debate here. How can we view these two distinct, in name, sciences especially considering how they developed interconnected but at separate times? My thoughts are that Mendelian Genetics was a first tier of scientific knowledge. Molecular Genetics is a transition to the second tier of S.K.

I think your question is quite pertinent, although a full answer would require a very broad and detailed account of reductionism. The benefits I perceive from reduction are frequently marginal. They can create a layered appearance of the sciences which would reflect what we perceive in nature (Water's discussed this in Thursday's lecture). But is this truly the reality of nature or is it merely our attempt to organize observations and place natural phenomena into categories?

Reductions also serve a unifying process in the sciences. Through this unification separate fields become related more closely which would appear to allow scientists with diverse backgrounds simpler means to work with one another. I would need specific examples of this simplification to uproot my skepticism of this. Is there not enough commonality between the sciences for cross-referencing without reduction? My skepticism sprouts from trying to explain consciousness or even the social sciences through reductions to physico-chemical processes. In the attempt to accomplish such a goal, a tremendous deal of phenomena would seemingly need to be disregarded, i.e. thought, belief, intuition, social dynamics, etc. Newly emergent fields like neuroscience may reveal laws about the physico-chemical formulation of such things which may allow for a reasonable reduction, but I remain skeptical that this is possible or a rational objective.

Another benefit from reduction originates from its failures. We all learn from our mistakes and so do the sciences. The failure of reductions can reveal dilemmas and procure insight into resolutions of these problems. This particular benefit I believe makes reduction a worthwhile endeavor. Even if there are no other benefits from reduction, its capability of distinguishing erroneous, contradictory, or incongruous dispositions of any given science makes reductionism indispensable.

This is a great question and I wondered the same thing. The concept of reduction did not come clear to me until I read Alex Rosenberg's "The Structure of Biological Science". From Rosenberg's perspective reduction can combine narrower theories into broader ones. It is also possible to reduce divergent and different theories into similar ones. So it just basically combines different theories into one complete theory.

It is harder for me to argue the worth of reduction as it is to admit to its inevitability. I think that reduction is a necessary and inextricable part of the scientific process and how it supposes to parcel, analyze and produce knowledge. If a general law or theory encompasses other laws and theories into a system of interrelation that is nevertheless more simplistic or unified than the diverse theories that it addresses, it would be dangerous not to reduce those theories to the general one. The question to me is not whether reduction is beneficial or not, but how does it function and what other forces are at play which may interact with or even mitigate reduction. Emergentism, perhaps?

I think the following may clear up why reduction is sometimes pursued in philosophy of science. At the very least, the following quote answers why taking a stance on reduction or at least thinking about reduction is important to philosophy of science. Consider the following quote from Ingo Brigandt and Alan Love in the SEP article "Reductionism in Biology." "Reduction is germane to a variety of issues in philosophy of science, including the structure of scientific theories, the relations between different scientific disciplines, the nature of explanation, the diversity of methodology, and the very idea of theoretical progress, as well as to numerous topics in metaphysics and philosophy of mind, such as emergence, mereology, and supervenience."

No matter where you fall in the reduction/antireduction debate, you are likely interested in it for the reasons explicated above. Asking whether reduction itself is necessary/useful/possible is no more than taking an anti-reductionist stance(see Kitcher for this). Though, I suppose you could think reduction would be useful and necessary if it were possible. If that were the case, asking why we care either way is hopefully answered by Brigandt and Love's quote. There are surely historical and sociological reasons for the appeal of reduction likely related to demands for simplicity and scientific unification.

I also had difficulty with the notion of "replacements" in science. It seems to me that most new theoretical frameworks in science incorporate at least some aspects of the frameworks that came before. I doesn't seem to me that wholesale "replacements" occur very often at all. I agree that attention needs to be paid to older theories in order to understand scientific progress.

In regards to your first question, I strongly agree with you. I find trouble with how reductionism can really help understand science better. Physics, mathematics, and chemistry are all utilized when studying genetics, but this doesn’t mean that the study of genetics is no longer a biological science. My ecology class I took years ago used as much calculus as my physics classes did, but this isn’t to say ecology isn’t a biological science. If the study of biology is constantly reduced to chemistry or physics, then what makes biology a distinct area of science?

As for your second question, I don’t know if I would argue that all scientific discoveries are revolutionary. First it must be defined what “revolutionary” means as well as “discoveries”. I know in my research projects it is not likely that our findings (discoveries) will be revolutionary in the sense that it will not dramatically change the way science is perceived or conducted, though this does not mean our work is useless. I research genetic causal factors in rare childhood cancers. The obvious goal for all researchers in this field is to cure cancer. Most of the research that will be conducted will provide clues and knowledge that will lead to a slow progression toward that goal. This is what I would deem normal science. It’s still very important science, since without all of this work, a revolutionary discovery would not occur. In this example, curing cancer would be the revolutionary discovery. I do agree with you in the fact that trying to define what is normal science versus revolutionary science is very subjective. I also agree with you about eliminating the distinction altogether because I don’t really know how this benefits science at all. My argument for different types of science was only to explain what Kuhn would have to say about normal and revolutionary science.

I also think that statement is false. I haven't taken genetics yet, but I am a biology student. In general biology courses, many aspects of genetics are talked about, such as chromosomes, DNA and RNA, genes, etc. Because of that, I'd say that genetics is still ruled by biology. I think it can be argued that physics and chemistry have been integrated into biology.
I'm also uncertain about the distinction between revolutionary and normal science. I think the distinction could be eliminated. Like you said, all scientific discoveries are revolutionary because they all essentially have a profound effect on the way science is perceived and passed on. Every discovery is important because you never know what will come from a discovery until time is put on it because of this the distinction between revolutionary and normal science could be eliminated.

I mostly agree with you. I find the concept of revolutions in science and politics problematic. Revolutions, whether political or intellectual are nebulous things those beginnings , ends, and causes are hard to define, those significance is hard to ascertain, and while they are considered to entail abrupt and discrete changes; in reality, revolutions do not occur in a vacuum. Thus revolutions are neither abrupt nor do they immediately result in discrete and identifiable changes. This is not to say that the zeitgeist does not change or that perspectives, world views, scientific theories are immune to modification; I am not suggesting that it is impossible to distinguish between pre and post ‘revolution’ instead I argue that these changes are slow and gradual. Therefore, these distinctions can only be made after the period of interest has elapsed. Contra Kuhn I do not think incommensurability is commonplace when scientific theories undergo changes. I think the idea of revolutions in science needs to, and has been, problematised.

How long can a scientific revolution take? When is a revolution completed? Doesn't these fetishism for revolutions lead to an oversimplification of the diversity of views that existed at any one time?

1) I disagree that genetics is 'ruled by biology.' If evolution is a process of natural selection - that is, a process that gives organisms with a higher fitness an increased chance of reproduction - then it would be contradictory to say that genetics is ruled by biology. If this were the case then the characteristics of a population would affect the genotypes of those organisms, rather than the genotypes dictating the characteristics of the population. Thus, I would argue that genotypes (molecular structures that change based on physical/chemical interactions at the molecular level) lead to biology, rather than biology ruling over genotypes.

2) I also disagree that all science is revolutionary. I agree that revolutionary discoveries profoundly affect the way science is done, but would you say that all discoveries truly have a profound impact on science? If you do a PubMed search on any popular topic, you'll find thousands of journal articles. Most of these discoveries could be called normal science - research that adds piece by piece to the overall database of scientific knowledge in a relatively slow manner. Chances are that most of these articles would deal with specifics that you are unfamiliar with. Alternatively, revolutionary science would be a discovery that has had a much larger impact - something that has drastically sped up the process of building knowledge and has had an effect on how we view the natural world. Therefore, I would argue that most science is in fact 'normal' science, and not revolutionary science.

I agree; the distinction between ‘revolutionary science’ and ‘normal science’ seems arbitrary and their use unnecessary. To me, it makes more sense to refer how much a science has changed as a result of new information/discovery. Attempting to determine if new angles on science constitute a revolution or not just appears to be a time wasting exercise as it pertains to advancing science and it may serve only as ego nourishment. But from a historical perspective the distinction between ‘revolutionary science’ and ‘normal science’ may be useful… but still arbitrary.

This is an interesting question because it deals with trying to eliminate the idea of reduction. I think for people who support replacement this would be an ideal, however I don't think its possible to do so since both are considered seriously by both sides.


I personally don't think that people try to reduce theories- it just makes the most sense at times. I agree that reductionism for the sake of reduction is silly, but sometimes it is easier to reduce a theory rather than all out replace it.

I think this brings up an interesting point within the dichotomy between development and inheritance that we've been grappling with over the last couple weeks. If we think that Mendelian genetics cannot (or at least hasn't been yet) reduced to molecular genetics, by reversing the flow of reduction, it highlights an intuitive notion that development and heredity are in fact one "thing"; we just have two different ways of viewing this "thing" that are currently not wholly inclusive. Perhaps these two lenses are in fact a dynamic of life in general; I mean specifically in relation to evolution, maybe it might make more sense to try to understand the units of evolution as interacting between these two dimension of the same process...