Question Submission 8

Please add your question as a comment to this post.

29 Comments

I am very interested in what Waters says in the last paragraph of section 3.4 of his book manuscript. (This paragraph starts "What's philosophically interesting about there representation structures...".) Generally, I would like to hear more about how Water's came to the conclusion that:
"Historians and philosophers tend to assume that the abstract models of science represent or are intended to represent the systems being investigated rather than the investigation itself or an evaluation of the usefulness of what is discovered. But centering our attention on practice reveals that classical geneticists devised different types of maps for each of these purposes."

Perhaps I am misinterpreting what is written but what exactly is it that implies that the scientific models developed by scientists were not just intended to represent the systems being investigated? I am tempted to believe that scientists had no larger intention than to convey what they had discovered with these models. This is not to say that I believe Water's is false, I just find this claim very intriguing and would like to hear more.

In his article "How Practical Know-How Contextualizes Theoretical Knowledge: Exporting Causal Knowledge from Laboratory to Nature," Professor Waters describes the "received view" of scientific theories. On this view, the study of artificially constructed cases in the laboratory identifies underlying universal laws which can then be applied to many situations beyond those initially studied. Waters rejects this view, at least as it applies to Morgan's findings. My question is: would Waters ever accept that experimentation on artificially constructed situations could reveal universal laws? Or would he contend that we can never know that these laws are universal and that, therefore, we should be conservative and believe that they only apply to the scenarios on which they have been tested? In other words, is there ever a point when enough experimentation has been done in a wide enough range of scenarios that we can proclaim a theory a "universal law"?

Chapter 3 of Prof. Water's book has a section on genetic that has me confused. I understand the initial purpose is to inform one the genetic mapping is alot more complicated than just finding the position of certain jeans. However, it seems like Water's is making the other three types of mapping seem more valid or more important. Are these types of mapping more important than the idea of being able to locate certain genes? And are these three other types of mapping as developed (useful) as genetic mapping is?

In chapter 3 of his manuscript, Professor Waters makes a case for the inherent problem with determining the physical relationship between different alleles on a chromosome via genetic linkage mapping through studding rates of recombination. The problem, as Waters reveals, is that crossover rate differs between different chromosomes and between locations on a particular chromosome. For example, the Morgan lab observed that the location of the allele for purple eye color experiences excessively high rates of cross-overs and the rate of double crossover is susceptible to temperature and age. For this reason, it seems that inaccuracies in the calculated distances between alleles are inevitable. My question is this: given the aforementioned problem with linkage mapping, and the fact that it poses a problem with determining the distances between alleles along a single chromosome, does the determination of the serial order of alleles along a chromos fall prey to the same problem; that is, does science still know the order of genes on a chromosome even though it appears that the speculated distances between alleles is flawed?

my question is about the wild type alleles. I know that this question arise in class but I stills have some doubts about it. The wild-type alleles is suppose to be the one that prevails among individuals in natural condition. Are this wild type alleles homozygous or heterozygous? And if Morgan lab have this fly in their lab and they control how they reproduce then they could control the wild type population, so then this wild type alleles are not the ones that prevails in natural condition. So my question then is what exactly is a wild type allele?

In Chapter 3 of Professor Waters' manuscript he states that, "The typical philosophical approach to analyzing a body of scientific knowledge places so much stress on theory, explanation, and modeling, that it obscures the nature of the body of scientific knowledge as a whole. I'm a little confused about what Professor Waters means by "obscures the nature of the body of scientific knowledge as a whole." Is it meant that philosophical approaches to scientific knowledge tend to disregard what the science was after but instead takes a step back to explain what the scientists did? Is the typical philosophical approach too keen on explanations and not on progress? Professor Waters also says that, "Philosophical accounts of classical genetics have focused on the explanatory principles and have not bothered to analyze even its most common investigative patterns…" It seems to me that philosophers and scientists should work together more so that problems like these don't arise. It's obvious that philosophy can help science progress and vice versa.

As a biology student, I am struggling to connect many of the philosophical concepts in Water's manuscript to present day research. He says in chapter 3 "Historians and philosophers tend to assume that the abstract models of science represent or are intended to represent the systems being investigated rather than the investigation itself or an evaluation of the usefulness of what is discovered. But centering our attention on practice reveals that classical geneticists devised different types of maps for each of these purposes."

This is all very important to understanding how the classical geneticists investigated and analyzed their research (and in this case, referencing genetic maps). How does understanding these older processes help in a practical sense with research today? How is understanding these philosophical concepts benefiting science in current times? I think that it is important for me to be able to make these connections because it helps me better understand the concepts.

In Waters "How Practical Know-How Contextualizes Theoretical Knowledge: Exporting Casual Knowledge from Laboratory to Nature", the second section discusses the "received model". He states, "because it relates to what has been called the received view of scientific theory...on this view, the exportation of theoretical knowledge from laboratory to nature is unproblematic because the knowledge gained in the laboratory is universal."

I would think that exportation of theoretical knowledge gained in the laboratory would not be universal. Later on in this paper, its stated that "differences in a gene cause uniform phenotypic differences in particular genetic and environmental contexts." It seems that the laboratory does not have the unlimited environmental contexts of nature, and thus could not be universal (since one environmental context could have an outcome completely different than another environmental context).

It is also stated that "in principle, if a theory truly accounts for the laboratory phenomena by identifying the underlying universal laws responsible for the laboratory phenomena, then it must represent the phenomena in nature as well." But is this "principle" really obtainable?

This idea just seems far fetched...so I'm wondering if people really still follow this kind of thinking?

In his paper, Professor Waters discuss the way in which genetic analysis was conducted. He discusses three important ingredients that go into successful genetic analysis, that being, transmission theory, procedural knowledge, and investigative strategies. Transmission theory we have discusses at length, but my question is how would Professor Waters define procedural knowledge? How does this differ from investigative strategies? What key distinctions make these both important parts of genetic analysis?

Professor Waters also discusses completed genetic analysis. He says that completed accounts typically “identified the alleles responsible for the mutant phenotype and the relevant regularities of expression.” What methods were they using to identify the alleles responsible for mutant phenotypes? Was it through phenotype observation or through other means?

In chapter 3 of Professor Waters’ manuscript, as well as in his lecture last Thursday, he discussed the importance of the maps created by Morgan and his team. I found it surprising that the constructional and valuation maps have been largely ignored by many philosophers and historians. In my opinion these maps are significant tools used in the investigative process. Being that I am neither a philosopher nor historian of science, I guess I don’t understand why such tools used in the scientific process have been left out of most analyses. When conducting an examination of science one would think all available tools, papers, data journals, etc would be taken into account and included in the analysis. So why have these been left out? Is this common practice in philosophical accounts of historical scientific processes? Is it just that philosophers (and historians) choose to focus solely on studying components that are theory-driven?

In section 3.3a Waters has a footnote stating that "scientists can employ an inference to best explanation within the context of an established body of knowledge." My question is why or how does 'inference to the best explanation' not apply to the explanatory scope? Classical genetics, as an established body of knowledge, certainly did infer a tremendous deal about genetic structure. From this explanation it could infer much about genetic inheritance, no? Can a science not infer the best explanation to broaden its explanatory scope? Is Waters claiming that inference is only true in application under specific parameters? And that inferences which reach beyond these parameters are inherently flawed?

In Chapter 3 - The Practice of Classical Genetics, Prof. Waters explains the several investigatory strategies not centered around explaining inheritance patterns. That is, those in the Morgan school were often after explanations of different phenomena such as chromosomal mechanics, map distances, the effects of new alleles and their relationships with other alleles, and the structure of the genome. To carry out this research, several sophisticated tests (which in themselves represent a significant portion of the scientific knowledge of the Morgan school) were used. These tests include the triple-backcross test, the complementation test, and tests and calculations for single and double crossover rates.

Though only a few of these endeavors resulted in genuine knowledge of the investigated phenomena, I am interested in whether or not they produced still something more. Beyond providing support for the transmission theory, does the success of certain investigative strategies provide some deeper theoretical knowledge beyond just the know-how of carrying out these patterns of reasoning? Isn't there some explanatory value in answering why these patterns of reasoning worked the way they did, assuming they were more than validations of the transmission theory, even if they could not be exported out of the laboratory? Could an answer of this type lead to better and more precise investigative strategies? Further, how would it do so? Even when considering the methods of material production, isn't there some theoretical or explanatory value in an answer of why certain methods resulted in sustained populations, or viable mutants? Of course, I am talking about the methods that were not just transmission theory-informed, if there were such methods. This question is not meant to cast doubt on the practice-based account of scientific knowledge. It is meant to enhance the view by claiming that perhaps this knowledge is even more valuable.

My question has to do with the linear arrangement of genes on the standard genetic map that represent the serial order of genes in chromosomes...
Waters says Morgan and Bridges "suspected that the crossover rate per unit of physical distance varies from chromosome to chromosome, and even from one region to another within the same chromosome."

I am more of less interested in how all this relates, as it most certainly should, to the fact that "Geneticists could not observe crossing over directly."

I find myself doubting the actual structure of this arrangement of genes in the chromosomes. I say this because they could not actually SEE the crossing over but could only observe phenotypic traits. I understand that they had other knowledge, such as chromosomal mechanics, that led them to make inferences about the recombination of genes.

B
U
T


Does it not make sense to consider the possibility that the most basic form of understanding in human knowledge comes in the linear form? And then we build to duplicate this knowledge and in the process break the original linear form... and recombine the knowledge to form an entirely new form of knowledge? (ie: DNA, RNA, polypeptides)
Does the structure of the chromosome and the genes located on it not represent just the thought process of the scientist(s) while they manipulated reality in order to produce results that went in accordance with their grander experiment? A potential cherry-picking of traits, flies and genes??

Just questioning...

"For this reason, it seems that inaccuracies in the calculated distances between alleles are inevitable."

I think your question is most interesting because I was traveling on a similar train of thought...

I feel that the inaccuracies in the calculated distances between alleles on the chromosome are inevitable as well. As I said in my question for this week, I have a problem with the fact that these scientist could not actually SEE the crossovers. They were making inferences about the knowledge. AND isn't this exactly what Waters said was "bad practice" of science... taking REAL knowledge of the world and expanding it to make inferences about other facts of reality. I understand that if you know enough about any system you can infer into the future a great deal with great accuracy. What I am suggesting is that these scientist might have wanted a clear cut answer so much that they engineered/ designed an experiment of large proportions such that they could use the "knowledge" gained in one part of the experiment to support another "knowledge" gained elsewhere. Were they not just institutionalizing? Limiting the results their labs reported as to avoid contradiction within the (united) institution?

I keep on going back to Morgan's paradox and the discussions of causality which surround it. In Chapter 2.4 Prof. Waters is discussing Woodward's Interventionist theory of causality and he makes the point that this theory is not reductive. Waters implies that this does not effect the utility of the theory, but that it may not be sufficient to please some metaphysicians or provide certain answers to what causes an effect. What would make a causal theory reductive? Are there theoretical examples, attempts or blueprints for such a reductive causal theory?

To me, all the elements of genetic mapping are equally important. Gene location is just as critical in understanding the process of genetic mapping as any other aspect. Perhaps, gene location is discussed less as it seems to be less controversial. It would be interesting to see a response from Professor Waters on why gene location is discussed less, but as I said, my guess would simply be that less confusion/controversy surrounds gene location.

Patrick's questions were answered in part by Prof. Waters in class. In short, Prof. Waters said that scientific knowledge is contextually situated in the way described by his paper.

But, I found Patrick's questions interesting if looked at in a different context. Let's imagine that the normal domain (classical genetics in our case) is advancing/changing through time, say, after the Morgan school to 1953 (Here, the discovery of the structure of DNA could be considered the starting point of the molecular revolution in biology). So, let's imagine that those parameters of which you speak have advanced enough to broaden the scientific context, and thereby broaden the applicability of the particular inference to best explanation. So, the explanatory scope of the science or its basic theory (I conflate the science and its basic theory, which may spell disaster, but this is just a blog post) seems to have increased along with the advancing investigative reasoning parameters. The question now becomes how does this change Waters' picture of the structure of scientific knowledge?

First, any additional theoretical knowledge cannot truly be accounted for without proper understanding of the investigative strategies and practical knowledge (context). So, the structure of that theoretical knowledge is still largely characterized by investigative strategy. But again, something has changed in the overall picture - the increased explanatory scope. Of course, Waters could say then that this is just part of the in principle part of the original explanatory scope. But, I think you raise a good question beyond that. My answer is that even if the explanatory scope broadens, it does so in an interwoven fashion with the investigative strategies. As those investigative strategies are contextualized across a broader range of investigative parameters, my prediction is that the investigative reach of the theory (in conjunction with all other components of scientific knowledge) also broadens. Then, the overall picture never really changes.

This is a prediction, and needs a vast amount of filling out. I also do not consider changes across Kuhnian scientific revolutions, because Waters account is interested in the normal science. We have to look at the actual history to see if this truly holds.

Being that I conduct scientific research, I too find myself questioning how reflecting upon past scientific breakthroughs applies to current research. The thing I find concerning throughout this course is the sentiment from our professors that philosophers often overlook much of the procedural components or process of scientific research, which I deem very important. As Professor Waters writes in Chapter 3 of his manuscript, "The typical philosophical approach to analyzing a body of scientific knowledge places so much stress on theory, explanation, and modeling, that it obscures the nature of the body of scientific knowledge as a whole.” Though I don’t dispute that philosophers of science have the potential to improve the scientific “method” and aid scientists through their analysis of science, I do question how often their research is actually applied to current science practices.

Based on my understanding of what the wild-type phenotype represents, it's not necessarily used as a link between what's found in the lab and what occurs in nature (i.e. when scientists use a wild-type phenotype, they aren't doing so in order to say that 'this is what occurs in nature,' but rather to say 'this is what happens in a non-mutant group in the laboratory').

While a wild-type phenotype is theoretically supposed to represent the most common form of the organism as found in the environment, it is difficult to have a laboratory model of an organism that represents the rest of its non-mutant species. When you look at modern scientific research, wild-types are used as a sort of positive control - a group that you can compare the experimental group with in order to see what your experimental manipulations effect. In doing this, the point of a wild-type phenotype is not to be able to relate your findings to organisms in the real world, but rather as a means to compare the experimental/mutant group to a control group within the laboratory.

I think it may have been that when the term "wild type" was first used, it referred to the type that is prevalent in nature. What is important about the "wild type" genotype is that it is a standard genotype used to compare against the mutation that is being studied. It is not always homozygous or heterozygous.

I believe that Waters is simply pointing out that the geneticists of the Morgan school had more than one way of mapping the locations of genes, and that all of these types of mapping reveal different knowledge. To map genes in a certain way provides evidence of the methods of the practice and different knowledge is required for each type of mapping.

It is tricky to come up with a concrete example of how Water's epistemology of scientific practice could or does directly effect contemporary scientific research practices. However, I think that the ramifications of his epistemology could effect certain important aspects of current research. At the moment, I can only imagine that most research money flows to strongly theoretically-defined programs. If Water's epistemology is taken into account it could alter the bias towards theory-centric research programs and convince investors to put their money towards more practice-centric endeavors.

I find this this question interesting: “Does the dtructure of the chromosome and the genes located on it not represent just the thought process of the scientist(s) while they manipulated reality in order to produce results that went in accordance with their grander experiment?”
It does seem that a linear assortment of genes along a chromosome as the explanation for the genetic arrangement of chromosomes fits well with how humans think. It seems to make sense to us—it seems even logical. But considering all of the manipulation and artificial selection that went into making the model organisms science use to study genetics (e.g., Drosophila melanogaster), who is to say this comes close to representing nature? The linear explanation of genes that science has fashioned may be a neat and tidy way of understanding genetics, but it may not reflect reality. The reality may be that genetics is far less ordinary from a human perspective.

I even want to go a few steps further (on a speculative tangent)—into a realm of the epistemological skepticism of the human ability to ascertain scientific knowledge. It seems that science makes the assumption that the human mind will be able to comprehend the mechanisms and processes dictating science. And, although the human mind is amazingly brilliant, there is no reason why we should assume this to be the case. Just like we don’t expect a cat to be able to understand simple addition, we shouldn’t assume that the human mind is capable of answering all of the various questions posed by science.

I think your question is pertinent. Even though all three maps were investigative tools, in my opinion, the standard map is the most readily comprehensible of the three. As a result, to explain what classical geneticists where achieving, historians and philosophers gravitated to the standard map as a means of simplifying and 'standardizing' explanation, which is to say the process by which historians and philosophers select the investigative tools to 'remember' is theory-driven; the tools which explain theory the most readily are the ones remembered. As time progresses, the gravitation grows in strength because more and more material becomes explained through this established method of analysis. Morange discusses this briefly in his introduction, how historians of science tend to focus on the 'stars', and the standard map would be the 'star' of these maps because of its relative accessibility. Unfortunately, this leaves valuable insight behind. I believe Waters is counterbalancing this phenomenon with his work. By evaluating the practice of normal science, he is incorporating the forgotten attributes (like the constructional and valuation maps) back into our understanding of normal science.

"Does it not make sense to consider the possibility that the most basic form of understanding in human knowledge comes in the linear form? And then we build to duplicate this knowledge and in the process break the original linear form... and recombine the knowledge to form an entirely new form of knowledge? (ie: DNA, RNA, polypeptides)
Does the structure of the chromosome and the genes located on it not represent just the thought process of the scientist(s) while they manipulated reality in order to produce results that went in accordance with their grander experiment? A potential cherry-picking of traits, flies and genes??"

In regards to your question- isn't this what every science experiment/new theory entails? Parts that are "unobservable", yet when explained in a matter support the rest of your theory? In another philosophy class, we discussed empiricism and its popularity in the past, as well as its complications. It seems to me that unless we are going to be absolute empiricists, we have to accept that THEORIES have some parts to them that aren't directly observable or explainable. Should we really knock work that isn't 100% observable or explainable?

The Questions: As a biology student, I am struggling to connect many of the philosophical concepts in Water's manuscript to present day research. He says in chapter 3 "Historians and philosophers tend to assume that the abstract models of science represent or are intended to represent the systems being investigated rather than the investigation itself or an evaluation of the usefulness of what is discovered. But centering our attention on practice reveals that classical geneticists devised different types of maps for each of these purposes."

This is all very important to understanding how the classical geneticists investigated and analyzed their research (and in this case, referencing genetic maps). How does understanding these older processes help in a practical sense with research today? How is understanding these philosophical concepts benefiting science in current times? I think that it is important for me to be able to make these connections because it helps me better understand the concepts.


This is a good question and already has some good responses. As Matthew already said, this is a hard question to answer. My contribution to the question is that we can hope that when better aware of what they are actually doing, scientists may be able to better alter their work so as to represent what they want to be doing. We cannot know how to improve the accuracy of our methods without knowing what are methods actually are.

I don't think Professor Waters is saying that one type of mapping is more valid or more important than the others. I think the idea of locating certain genes is just as important as the other types of mapping. Different things can be found out from the different types of mapping. Each type of mapping is useful, depending on the kind of information you want to find out.

I guess what you are hinting at gets down to the question of what you, because at this basic level it is a personal belief, believe the function of science to be...


I am more interested in understanding reality as human's ability allows them to understand it and explaining WHY we cannot understand the portions that we do not understand.

I am far far far less interested in knowledge that "covers" up the flaws of human understanding and in fact allows human's perception of human knowledge to be falsely advanced.

I hope this makes my previous statements more easily understood.

I don't think that Professor Waters would accept that experimentation on artificially constructed situations could reveal universal laws. After listening to him lecture, I believe that he would be conservative and believe that the laws only apply to the scenarios on which they have tested. Even when making an artificially constructed situation, not all of the possible environmental contexts are going to be able to be introduced (because they are unlimited) and thus the law can not be universal. There are always going to be exceptions which would lead to a law not being universal.