Question Submission 14 (Last One)

Please add your question as a comment to this post.


Might it be the case that theories themselves have aspects of intrinsicality, fundamentality, and temporality, much like reduction between theories does? There are clear differences in how theories work, especially between physics and biology, and it is clear after reading Professor Love's paper that this affects how reduction is done. But is this a function of how the reductionist approaches and describes the theories involved, or is it something about the theories themselves that determines when a conflict will occur? I guess what I am really trying to get at is whether it is possible to get a reduction between physics and biology that does not violate one of the three aspects, or do the content or goals of the theories make such a pure reduction impossible?

First, I believe the shift between molecular and classical genetics is marked by a much thicker line than pre/post double helix, though I do not think Waters would disagree (no sharks here, I just need this information to help set up my question). The advent of biochemical genetics, through the work of Beadle and Tatum, seems to mark an important pre-double helix shift toward molecular biology. What unifies Waters' account with this picture, is that I take Beadle and Tatum's work to have been based around a practical investigative advance/success. Not only that, but this practical shift is nothing more than a retooled version of the genetic approach made popular by the Morgan lab.

So, my question is that given what I take to be a fairly identifiable/stable investigative strategy (the genetic approach), which has slowly evolved over the history of genetics , can we classify it as generatively entrenched element in the knowledge structure of genetics? Would being such a structural element give justification for philosophical attention when constructing epistemologies of these domains, especially given its influence on the practice? Lastly, and this seems the most difficult to answer, is the genetic approach's generative entrenchment the result of its success? Certainly not all entrenched elements are generative, nor are all generative ones so because of their success.

I appreciate the pragmatic way which Love and Huettemann tackle reductionism. Up to this point it had seemed as though the discussion regarding reductionism was polarized in black and white terms between the reductionists and the anti-reductionists. Accordingly, the new facets of reduction that they discuss – intrinsicality, fundamentality and temporality – strike me as being reasonably fluid and practical. Alongside this aspect of fluidity, however, comes the possibility that these terms are being determined by contemporary practice in a negative way. For example, with intrinsicality in mind, Love and Huettemann say that it is the researchers role to define what constitutes an intrinsic system. They need to create (ambiguous or even artificial?) boundaries that make sense for the goals and scope of their research, but these boundaries do not necessary exist in the clear cut way that they are made to for the nature of that research. These boundaries would and are set up outside of concerns regarding reductionism, but is the veracity of these distinctions sometimes harmful for reductionistic and other considerations. Is it really safe to begin conjecture from the belief that a neuron, for example, is an intrinsic system? Is this a necessary risk of tackling reductionism from the perspective of practice?

In Prof. Waters's "Beyond Theoretical Reduction...." I was interested by his talk of theoretical reasoning and investigative strategies that advanced the investigative powers by producing new methodologies. Furthermore, he talks about how this increased the investigative utility and is now utilized in various disciplines of biology.
My question is:
How can we understand the origin of new divisions of science that emerge out from the use of new methodologies? Does the development of these new divisions of science from the use of the same methodology(ies) constitute a level of organization within the main science itself?
Finally, it looks to me like that is where a reduction could start. But I think this would lead to only a partial reduction. I say this because I think there is a lot more to special divisions of, say biology, than what they contribute to biology as a whole.

In Prof. Waters’s "Beyond Theoretical Reduction and Layer-Cake Antireduction” he explained Rosenberg idea that the relation between molecular genes and Mendelian phenotype is too complex to make a connection between these concepts. My question is, this is not what the scientific community is trying to do? Trying to find an explanation of phenotypic traits by looking in molecular genetics? Is this not trying to reduce classical genetics to molecular genetics?

After reading Prof. Waters’ ‘Beyond Theoretical Reduction and Layer-Cake Antireduction,’ I’m having difficulty in piecing together the two main discussions. The first discussion deals with how the two opposing views of theoretical reduction and layer-cake antireduction go astray – Waters concludes that attention should be shifted from these theoretical differences to investigative practice and that focusing on the better explanations that reductionism may offer ‘is not the development in genetics that has transformed biology’ (p. 250). The second discussion deals with how Watson and Crick’s discovery impacted genetics and transformed practice in many biological sciences – Waters concludes that the post-Watson and Crick development of genetics ‘involved a new basic theory that recast ideas of classical genetics in terms of molecules’ (p. 258) and that ‘the recasting of the basic theory is important because it makes it possible for biologists to retool and build upon the basic investigative approach of genetics’ (p. 258).

Waters states that ‘attention should be shifted from theory to practice’ (p. 239) and that the retooling, not the theoretical aspect, has been more transformative to biology. To me, these seem like two separate and legitimate areas of discussion. One the one hand, we have the debate between theoretical reduction and layer-cake antireduction – what I take to be as points of view that differ in how higher- and lower-level information should be interpreted. On the other hand, we have a discussion about how biological sciences are now able to use genetics as an investigate tool to study various phenomena. I’m having a hard time viewing these two discussions as directly inter-related – that the focus can or should be shifted from one to the other. Is it appropriate to view these implications (theoretical and practice-based) as separate entities that can have their own domains of discussion – each viewed with the amount of focus one desires? Or is it more appropriate, as Waters argues, for attention to be shifted from theoretical to practice-based implications? What would be the problems associated with viewing each as a separate entity, and how might a shifting of focus resolve these issues?

Professor Waters states that, "The chief arguments offered in favor of layer-cake antireductionism in genetics fail to correspond to the actual science." The reasons pro layer-cake antireductionists give for the view that classical genetics will never be reduced, eliminated, or explained away are based on assumptions. The argument that the two tiers of discourse correspond to classical and molecular genetics can't be systematically connected because claims made in gene terms can't be connected to claims couched in terms of DNA is an, "assumption that, in classical genetics, the relationship between a gene and a phenotypic trait is taken to be simple." It's obvious with what Professor Waters states in his essay that layer-cake antireductionists are wrong, but layer-cake antireductionists are still around. It seems to me, philosophers that have spent a lot of time and effort in the field of classical genetics don't want to admit that their science can be reduced because in a way it says that what they have worked on all those years isn't as important. Why is the continuation of debating amongst philosophers so central to what they do? Couldn't a lot more progress be made if philosophers could admit when they're wrong and just move on to advance the sciences. Isn't that, in the end, the most important part about science? I'm also a little confused about scientists and their view on this debate. Philosophers are interested in the debate about whether classical genetics can or cannot be reduced, but are the actual scientists that take part in classical genetics interested in the same debate?

A layered cake view of reductionism seems compatible with a different view of reductionism - layered reductionism (if you will). What I mean, is that one can oppose full scale reductionism and embrace intra layer reductionism (seeking to explain phenomena within a layer by the simplest or most 'fundamental' pieces at that particular level. Nevertheless, this still seems to pay too much attention to theories. If, like Waters suggests theory follows practice then those things that are 'reduced' appear to have been reduced as a result of successful manipulation. In effect, is the explanatory range of a theory a collection of the phenomena that has been reduced to that theory?

Waters explains that the cardinal argument in support of layer-cake antireductionism and the idea that “classical genetics will never be reduced, eliminated, or explained away because its central theory explains kinds of phenomena that are best explained at the level of classical genes and chromosomes” is flawed and that it fails “to correspond to the actual science”. I am in agreement with Waters on the fact that layer-cake antireductionism ignores an abundance of scientific evidence when it is suggested that “classical genetics will never be reduced, eliminated, or explained away” but I do disagree with the notion that some phenomena are not “best explained at the level of classical genes and chromosomes” and are so at the lower levels of biology. I believe that there is a lot of useful, practical potential buried somewhere in the layer-cake antireductionist perspective as long as one does not make the mistake of rejecting reductionism.
Isn’t the layer-cake antireductionist on to something? Doesn’t layer-cake antireductionism provide insight not available to a reductionist account of biology, at least in a practical sense?

I think that most theories or practises that become entrenched do so because they are perceived to be successful. I would argue that there are two ways this can happen - one they can actually be successful or they may hitchhike along other elements to which the success would be rightfully attributed. However, it seems that once a theory has had some success within its proper explanatory range - practitioners forget the initial boundaries that enabled the theory to be generative and wrongly enlarge its explanatory range.

In Professor Love’s article I found myself very interested in his discussion of temporality. Thinking about my area of expertise as a cancer researcher, I think about the methods of how cancer research is conducted. It makes more sense to take a temporal approach when researching cancer since the system at T2 is causally related to mechanisms that occurred at T1 and cancer is a disease that develops over a long period of time. But I would assume that cancer treatments are focused more atemporally, seeing that they focus on treating the parts at T2, when a tumor has been identified, though the parts at T1 and T2 would be at different levels (the parts at T1 study mutated genes, whereas the parts at T2 focus on the tumor cells). Does this seem like an effective approach to take; having a disconnection over which time period the parts contributing to cancer are focused? Seeing that most contemporary cancer research methods take a reductionist approach, maybe they should consider subsequently studying cancer at a higher level as well?

I found Wimsatt's lecture on Thursday to be interesting and thought provoking. When lecturing on how to identify a reductionist- I found myself being described perfectly. My question is the criterion of multi-level reductionistic modeling. I feel as though many reductionists aren't as type A as this criterion makes them out to be. Aren't there cases in science where "going up and down" isn't really possible due to lack of "scientific evidence" yet people still claim reduction?

Professor Love discusses the negative view held of reductive theory in the biological sciences. He proposes that we alter the way reductive theory is used to create a more useful, meaningful process. He says the focus should be more on actual scientific reasoning and highlights the three aspects (intrinsicality, fundamentality, and temporality) that come from composition and causation. He then continues to say that this will offer new forms of reduction that will either succeed or fail at adequately reducing/explaining biology. My question is, have I properly understood that Professor Love feels this alternative view of reduction will help measure failure or success? If so, what would Professor Love consider to be a successful reduction?

I think this highlights much of our discussion this semester. When is reduction appropriate? When is it useful? Is it okay to due even if the scientific evidence is diluted in the process? Great, concise question.

Waters' discussion of theoretical reduction and the layer-cake anti-reduction is provocative. It seems to me that Waters' demand for shifting towards a practice-centered interpretation of knowledge would be a natural result of evaluating reductive theories with Love's three aspects of theory (intrinsicality, fundamentality, and temporality). Is this an improper judgement? I feel each of these aspects would necessarily be determined by a science's practice and thus should manifest in its' theory. And even furthermore in a reductive theory between fields. I couldn't ascertain why this connection would fail. It just popped into my head and wouldn't leave. Hopefully one of you can explain why this connection is true or false. Or maybe just marginal.

I agree that biological scientists are investigating the generation of phenotypic traits at the molecular level. But I do not believe they are trying to reduce classical genetics to molecular genetics. The scientists are looking for more accurate and coherent explanations of these phenomena, and biochemistry is currently an encouraging methodology. Attempting to improve biology's explanatory scope does not necessarily mean reducing it to chemistry. Whatever honestly and definitively improves explanation will be accepted. And the reduction will follow if it is plausible. Waters' mentions that scientists aren't typically investigating to further reduction. Reduction occurs as a result of organizing the explanatory scope of various fields. At least that is how I understood it.

I think the last question you had perfectly describes why reduction fails, the majority of the time it does. The cases that I can imagine in which going up and down is difficult, because of the lack of scientific evidence, all seem to be sciences (divisions of science) that are too far separated from one another for a reduction to be successful. I believe that reduction can always look good at a glance but once you get into the details it gets a bit tricky.
To get back to your last question: "Aren't there cases in science where "going up and down" isn't really possible due to lack of "scientific evidence" yet people still claim reduction?"
I want to comment on 'yet people still claim reduction' here. I think it is important to maintain the distinction between someone claiming reduction and someone claiming a possible reduction. I say this because if someone claims a possible reduction they may make this claim under a certain context. In this context they may, knowingly, not legitimately understand where the science they are interested in will end up going. AKA what research will continue? what new research will surface? what will draw in funding? Anyways, this someone may claim a possible reduction, in which they outline how a reduction could take place from one science (division of science) to another. They may however not necessarily believe that this reduction will actually take place.

Some good thoughts...


I definitely think that the explanatory range of a theory is a collection of the phenomena that has been reduced to that theory. Something like DNA replication is a good example to me where the theory of it is drawn from each individual phenomena or step in a molecular biology sense. I think there are many other examples in this field that play into exactly what youre talking about.

Thanks for the response Chris. I'm interested in what you say about theories, mostly that they can be inappropriately applied outside of their explanatory range once they have become successful (and thus entrenched) within their proper explanatory range. First, let's assume that there is a well defined "proper" explanatory range. Stepping out of the proper explanatory range, if done carefully (i.e. using Wimsatt's false model heuristics), can expose the shortcomings of theory and hopefully lead to modifications in that theory that are suitable for explaining the new phenomena. If not, then a new theory could be born out of the failure of the old. If we assume that some portion of the original theory is left intact (we don't wholesale reject it) are the remaining elements considered to be generatively entrenched within the new range? That is, they generate newer explanatory elements, that are dependent on the accuracy of the original elements. Isn't this a viable path towards GE?

Of course this is a hypothetical example. But, if possible wouldn't this be a way to appropriately apply a theory outside of its explanatory range, which results in its being entrenched within a new explanatory range? Or, if in a Waters-esque epistemology, the use of a theory as investigative tool to probe aspects of the natural world beyond the range of the theory becomes an generatively entrenched investigative tool rather than explanatory element? Additionally, it seems that using tools like genetic analysis can only become GE if they are applied outside the original domain in which they were intended.

At first I had the same intuitions as you, that we could have separate discussions of these matters (reduction and theory vs. practice centered views of science), but I believe that there are two things that prevent this. First, reductionists and antireductionists both claim to have an explanation for what happened in biology after Watson and Crick’s discovery. This forces the two discussions together, even though we can imagine a situation in which these are simply views about reduction and nothing more. The facts prevent a separate discussion. Waters is saying that because the focus on how biology has transformed over the last 60 years has been focused on reduction, and has been theory-biased to the point of disregarding practice, the real element of change has been missed. If Waters is successful in turning the discussion away from theory, then the antireductionist/reductionist debate does become purely about how the science is structured, but as is it is part of the debate about how the DNA model changed the science.

A good question. I have kind of been wondering the same thing. Hull also talks about the reduction of Mendelian genetics to molecular genetics and in my opinion Mendelian genetics and molecular genetics aren't different, but rather just different perspectives of looking at the same thing.

The idea of seeking an explanation of a phenomena within a layer with the simplest or most fundamental pieces at a particular level is intriguing. However, I think you're right that it does seem this view would pay much attention to theory (taking a theory and looking at it from the particular level). I also think that the explanatory range of a theory is indeed a collection of the phenomena that has been reduced to the theory.

I agree, to me it seems that modern science is engaged in an effort to reduce classical genetics, but the thing is that this is more of a result of the current direction of science and not a main goal of modern science; that is, the aim of current science is not to determine if reductionism of layer-cake antireductionism is applicable, it is to yield sound data that produces practical results. To go off on a tangent, in my experience, many scientists would likely report an agnostic position as to whether classical genetics is reducible to something else or not. Science, modern science at lest, is rooted in practice and results while retaining the realization that their knowledge of nature is limited, caring less about every single little detail of how some process works then about those details that are relevant in a practical sense. Take medicine for example: it is useless to know that a certain set of parameters is beneficial at harboring a specific condition if there lacks a realistic potential of producing such parameters in a living subject. There is a point where academic philosophical analysis fails to benefit science and for this reason science and philosophy of science should remain two distinct endeavors, in my opinion.

"Does this seem like an effective approach to take; having a disconnection over which time period the parts contributing to cancer are focused? Seeing that most contemporary cancer research methods take a reductionist approach, maybe they should consider subsequently studying cancer at a higher level as well?"

This is an interesting question because it does a nice job of tying in-class concepts to real-world problems. Addressing your question of different temporal approaches, I think that's the best option currently available. Research is being carried out to understand how T1 results in T2, but until those answers are found the best strategy, in my opinion, would be to treat each condition separately since we don't fully understand the causal mechanisms that link them together.

Addressing your question of higher-level approaches, I'm not sure if I can think of any ways in which that would be more advantageous than the reductionist approach. Cancer, to the best of our knowledge, is a molecular phenomenon. Most of the research goes on at this level because this is where we think the causal mechanisms are taking place. Taking a higher-level approach seems more appropriate at T2 and after, once the molecular events have caused higher-level changes. Of course, there are examples of practices (e.g. those that deal with the hereditary nature of cancer) that use higher-level approaches, so I don't think it's been avoided entirely. But from the treatment aspect of things, I think the reductionist approach is the most logical at this point in time.