RESEARCH PROJECTS

The following thirteen projects are at different stages of development and represent a variety of ongoing research (although not all of it). Some of them arise out of my doctoral dissertation (see description below; scroll down), whereas others are new endeavors. Please contact me for further information (aclove@umn.edu).

(i) Reductive Explanation and Temporality

Time is an important factor in biological explanations and one that has not received sufficient attention in the context of reductionism. This project focuses on epistemological aspects of time and their relevance to reductive explanations, especially as seen in cell and developmental biology. These analyses are part of a larger investigation on representation and explanation in contemporary developmental biology.

* * *

(ii) The Problem of Innovation and Novelty and Integrative Explanation in Biology (with Ingo Brigandt)

This project attempts to extend my earlier published research and dissertation analysis of the concepts of evolutionary innovation and novelty, especially the sense in which research on evolutionary innovation and novelty require interdisciplinary collaboration and how this research constitutes a distinct domain of investigation separate from adaptation in evolutionary biology. It also links up with a collaborative endeavor with Ingo Brigandt entitled "Integrating different biological approaches: a philosophical contribution", which is currently being funded by a Canadian Social Science and Humanities Research Council grant.

* * *

(iii) Asa Gray's Evolving Perspective on Teleology, Variation, and Natural Theology

Asa Gray was a close confidant of Charles Darwin but had differing views on religion and natural theology. Although scholars have recognized that Darwin's ideas change over time, Gray's own thinking has not been viewed through the lens of conceptual change. His views on variation, teleology, and natural theology are quite different later in life than often portrayed. This is part of a larger research project in collaboration with John Beatty and Jim Lennox that dissects the 19th century conversations on chance, purpose, design, and mechanism by Darwin, Gray, and others, and explores its significance for contemporary analyses of related concepts in evolutionary biology.

* * *

(iv) Formal versus Material Theorizing in Philosophy of Science

John Norton has argued that there are no universal rules of inductive inference. In his preferred ‘material theory of induction’, the license for particular inferences are local facts and therefore formal theories of induction that abstract away from this empirical content, with the goal of being generally applicable to any science and all inference, are flawed. Although disagreement remains about Norton’s substantive thesis, his methodology of pursuing material rather than formal theories in philosophy of science can be generalized: the license for all aspects of scientific reasoning is local (or at least not global), whether discovery, confirmation, explanation, or theory structure. But by itself this perspective is problematic because it does not recover why formal approaches have had so many successes, even if they have not been universal. This project strikes a balance between formal and material theorizing in philosophy of science in order to explicate the actual procedures of scientific reasoning and yield more accurate insights about its operation. The result is a more heterogeneous conception of philosophy of science that is commensurate with its sizable disciplinary growth over the past several decades. It also demonstrates that some issues that exercise philosophers are pseudo-problems (e.g., Fodor's complaints about evolutionary biology), not because formal approaches are unhelpful but only because they are not universal.

* * *

(v) Sciences without Theories?

In Science Without Laws (1999), Ron Giere observes that most theorizing about science has been colored by a set of concepts that held special importance in the modern period: laws of nature, scientific truth, and scientific rationality. Pictures of science focused on these concepts came under severe scrutiny through the work of Kuhn and others, which emphasized how the practice of science did not conform to these categories. “Science need not be understood in these terms and, indeed, may be better understood in other terms” (Giere 1999, 4). Another concept that has played a central role in philosophical analyses of science is ‘theory’: How are theories confirmed? How do we decide between competing theories? How are observations biased by theories? Are theories underdetermined by evidence? What is the structure of scientific theories? It is difficult to imagine how one might proceed in the absence of theory: can some sciences be understood in other terms? In this project I am constructing an affirmative answer to this question by focusing on problems as organizational resources that can substitute for the assumed role of theories as systematic representations of scientific knowledge. In particular, I highlight three neglected features of scientific problems: hierarchical structure (‘depth’ or ‘complexity’), interdisciplinary location, and diversity of kind (empirical, conceptual, semantic, etc.). Using these three features I drawn an analogy with anatomical structure, which emphasizes the material (rather than formal) structure of problems; i.e., the structure is specific to the content of the sciences where the problems are manifested. Several advantages emerge from this approach, including a straightforward explanation of exploratory experimentation and a novel interpretation of incommensurability.

* * *

(vi) Modeling Experimental Evidence from the Practices of Developmental Biology

Many philosophical analyses of confirmation or hypothesis testing proceed on the assumption that the evidence has been gathered already. These ‘after-the-fact’ treatments miss important features of scientific evidence because determinations of relevance and the inferences licensed often depend on the circumstances of procurement. Philosophers must attend to the details of these scientific practices in order to explicate how these judgments of evidential relevance emerge out of experimental inquiry. Some philosophers have explored the procurement of evidence under the label “data models”. Unfortunately, they only address statistical evidence, encompassing the process in two steps: (a) data reduction (error elimination); and, (b) curve fitting (‘clean’ presentation). This overlooks key aspects of the evidential practices observable in experimental biology, especially those that involve images. To redress this lacuna and better understand pictorial evidence, I am exploring the practice of producing images of gene expression patterns from in situ hybridization experiments and how they become evidence in developmental biology. The practice of pictorial evidence production can be likened to modeling, and the resulting images can be considered models. Therefore we can analyze this practice by looking at “modeling choices” such as idealizations—knowingly ignoring variation in properties or excluding particular values for variables. Because many of these choices are made long before a scientist executes a particular experiment, I introduce the notion of “serial idealization” to capture this temporally extended practice and distinguish it from other senses of idealization. The practice of serial idealization can be characterized in terms of three phases (upstream [model system, research problem]; mid-stream [process in view]; downstream [particular experiment]) and three sources of origin (forced [unable to model unless choice is made]; entrenched [past choice established in the research community]; conventional [choice could be made otherwise but is not]). Many upstream and mid-stream choices that are entrenched or conventional can be found embedded in experimental practice. As a result, discussions of confirmation that focus on forced downstream choices systematically ignore the idealizations that illuminate how evidence comes to bear on hypotheses. My analysis of the practice of serial idealization: (1) explains how different disciplines using the same experimental methods can disagree over standards of evidence (including artifacts) and maintain biases in data gathering; (2) enriches our understanding of data models because pictorial evidence is not akin to curve fitting where an indirect summary representation (graph or histogram) is utilized; and, (3) expands our notion of scientific modeling to include more than the “theoretical” approaches that have been a mainstay of philosophical discussion.

* * *

(vii) A Pluralist Stance on Top-Down Causation

The ubiquity of top-down causal explanations within and across the sciences is prima facie evidence for the existence of top-down causation. Most scientific and philosophical debate has focused on the status of the latter: is top-down causation coherent? Is it in conflict with reductionism? Less attention has been given to three assumptions involved in deriving an account of top-down causation from top-down causal explanations found in the sciences: (1) the representations of hierarchical relations (i.e., tops and bottoms, and their temporal relations) are consistent across different sciences; (2) the meaning of causation is the same in all of the sciences; and, (3) scientific explanations should be understood causally rather than in terms of composition, deductive argumentation, or appeals to law-like relationships. I argue that the diversity of scientific practice demonstrates that all three assumptions are violated: representations of hierarchical relations are not consistent across different sciences; there are distinct meanings of causation in different areas of science; and, explanation is understood in more than causal terms. I illustrate these points with three corresponding examples: (i) non-coincident spatial decompositions of organisms into parts by different biological sub-disciplines; (ii) causation as invariance of properties under intervention in biological science versus causation as events governed by laws in physics; and (iii) (bottom-up) compositional versus (top-down) causal explanations of protein folding at the junction of biology and physics. The violation of these three assumptions suggests that no single, unified metaphysical account of top-down causation emerges from the epistemology of explanation in scientific practice. Instead, we should adopt a pluralist perspective on top-down causation—there are many different kinds of top-down causation (explanation) and they are consistent with the simultaneous existence of many different kinds of bottom-up causation (explanation). This has the added advantage of making precise why several different senses of top-causation are coherent and why they are not in conflict with reductionism.

* * *

(viii) Reductionism and Evolutionary Explanations of Morality

One common kind of claim (implicit or explicit) is that morality can be reduced to or explained away by evolutionary biology. What is involved when we try to reductively explain morality with evolutionary biology? How will we know whether the project has succeeded? Does explanatory success imply an elimination of morality as currently understood? What are the criteria of adequacy that form the necessary prerequisites for evaluating evolutionary explanations of morality that claim to be reductive? Reductionism is always a claim about two domains: a domain to be reduced or explained, and a domain that reduces or explains. These two domains can be within an area of science (e.g., biology: reductively explaining classical genetics with molecular genetics), across areas of science (e.g., physics and chemistry: reductively explaining chemical bonding with quantum mechanics), or from a non-scientific or everyday domain of discourse to a scientific domain (e.g., biology and religion: reductively explaining religious beliefs with neurobiology). We can codify these differences in a three-fold distinction: reductionism 'in', reductionism 'across', and reductionism 'to'. Reductionism 'in' biology involves relating domains of different disciplinary approaches in the life sciences, such as reductively explaining animal behavior by appeal to brain anatomy or hormonal signaling. Reductionism 'across' sciences involves relating domains from different areas of science, such as reductively explaining cardiac anatomy (biology) by appeal to fluid forces in blood vessels (physics) during embryonic development. Reductionism 'to' a science involves taking some feature of a domain that is understood to be non-scientific—religious beliefs, moral judgments, political persuasions, or musical preferences—and reducing it to one or more sciences (e.g., behavioral ecology, neurobiology, or biochemistry). The subject of reductively explaining morality with evolutionary biology is in the realm of reductionism 'to'; reducing a non-scientific/everyday domain (morality) to a scientific domain (evolutionary biology). The success (or failure) of reductionism 'in' or reductionism 'across' depends on the explanatory standards in these domains of scientific inquiry. The introduction of reductionism 'to' complicates the situation because a domain of scientific inquiry and some other discourse that is usually understood as non-scientific must be related to each other. Two questions are critical for articulating these explanatory standards: (a) What epistemic units are used to represent higher and lower-levels in the reduction? (b) How should we interpret the “reduction” of higher-level representations to lower-level representations? Answers to these questions help delineate the license for successful reductive explanatiosn. Therefore, we need to spell out criteria of adequacy for reductive explanations in biology (reductionism 'in') to ascertain the interpretive possibilities for claims that morality can be reduced to or explained away by evolutionary biology (reductionism 'to').

* * *

(ix) Noise and Information in Biological Science: A Study of Conceptual Behavior (with Kyle Menary)

One of the most widespread and yet controversial concepts in contemporary biology is information. Recent studies of conceptual behavior by philosophers and psychologists have indicated a previously unrecognized complexity in the use of concepts. This projects attempts to get at this complexity by looking at a concept that necessarily coordinates with information: noise. Cell and developmental biologists routinely use 'noise' in investigation and explanation, which reinforces the view that conceptual behavior science requires analyzing concepts in conjunction with one another rather than in isolation.

* * *

(x) Conceptual Change

How do scientific categories and their corresponding terminologies mutate and evolve over time? Following the work of Kuhn, much of the discussion about conceptual change has focused on ‘incommensurability’, understood as the existence of different meanings for the same terms used in competing paradigms that cannot be adjudicated by rational discussion. Various attempted solutions to this situation based on causal theories of reference have left conceptual change relatively opaque and ignored its diversity: the introduction and elimination of concepts, a reclassification of things considered to fall under a concept, the development of more abstract concepts, the refinement or expansion of ‘defining’ features of a concept, and the reorganizing of relations among concepts. This research project attempts to offer a new perspective on conceptual change in terms of concepts being jointly deployed in scientific explanations and draws attention to a neglected perspective on incommensurability: the lack of a common measure between problems, data, methods, or criteria of explanatory adequacy that arises when articulating ‘integrated’ explanations. The philosophical account also directly addresses key epistemological issues in philosophy of biology, such as determining when explanations are competing or complementary.

* * *

(xi) Actualism, Historical Science, and Paleobiology (with Lance Lugar)

Explaining the Cambrian Radiation has been a key issue for evolutionary biologists all the way back to Darwin.  Recent literature reveals key differences in the way biologists use the philosophical principle of actualism.  Actualism underlies Lyellian uniformitarianism but by itself makes no commitment to the uniformity of causal intensity in the history of life. It only prescribes the use of causes now in operation to explain historical events in the past.  In this paper we dissect the arguments over how to explain the Cambrian Radiation and diagnose part of the explanatory conflict in terms of assumptions concerning the principle of actualism.  In short, researchers from a particular specialty are willing to relax the principle for their domain of expertise, but not others.  This shows how subtle differences in philosophical assumptions can lead to major difficulties in assessing the value of competing scientific explanations, which connects with recent literature in philosophy of biology about explanatory pluralism.  Our discussion of actualism also allows us to comment critically on recent articles that explore the differences between experimental and historical science.

* * *

(xii) Thought Experiments in Philosophy and Philosophy of Science (with Megan Delehanty)

Conceptual analysis in contemporary philosophy relies on the use of possible cases to evoke intuitions that can be used as ‘evidence’ for or against philosophical theories in particular domains.  This type of analysis has come under increasing attack with respect to the problems that attend exotic fictionalized scenarios and the static notion of concepts that is often assumed.  In this project we offer a general framework for being skeptical of this approach and identify its problematic deployment in philosophy of science where it is somewhat masked amidst a welter of interesting scientific detail.

* * *

(xiii) The Phylogenetic Approach to Philosophy of Science (with James Lennox)

Twenty-five years ago there was a vigorous discussion about the relations between history of science and philosophy of science.  This topic warrants revisiting now that philosophy of science has experienced a fruitful trend of specialization by attending to the specifics of different particular branches of science.  The status and use of history is still a major issue that divides contemporary philosophers of science.  We address the nature of the epistemic currency focused on by philosophers of science (concepts, explanations, models, theories, etc.) and articulate a rationale for a philosophy of science that focuses upon foundational problems in particular sciences that simultaneously requires attention to the historical origin of the problems in order to diagnose the philosophical difficulty, and potentially prescribe a solution.  The approach is demonstrated through our own case studies and those of other researchers exhibiting this mode of inquiry.  Our goal is to reinvigorate this discussion among philosophers of science by providing new arguments in favor of a critical role for history in philosophy of science.

 

* * *

Dissertation

Explaining Evolutionary Innovation and Novelty: A Historical and Philosophical Study of Biological Concepts (PDF version)

Committee

James Lennox [Director] (History and Philosophy of Science/Center for Philosophy of Science - University of Pittsburgh); Sandra Mitchell (History and Philosophy of Science - University of Pittsburgh); Robert Olby (History and Philosophy of Science - University of Pittsburgh); Rudolf Raff (Biology - Indiana University, Bloomington); Günter Wagner (Ecology and Evolutionary Biology - Yale University)

Abstract

Explaining evolutionary novelties (such as feathers or neural crest cells) is a central item on the research agenda of evolutionary developmental biology (Evo-devo).  Proponents of Evo-devo have claimed that the origin of innovation and novelty constitute a distinct research problem, ignored by evolutionary theory during the latter half of the 20th century, and that Evo-devo as a synthesis of biological disciplines is in a unique position to address this problem.  In order to answer historical and philosophical questions attending these claims, two philosophical tools were developed.  The first, conceptual clusters, captures the joint deployment of concepts in the offering of scientific explanations and allows for a novel definition of conceptual change. The second, problem agendas, captures the multifaceted nature of explanatory domains in biological science and their diachronic stability.  The value of problem agendas as an analytical unit is illustrated through the examples of avian feather and flight origination.  Historical research shows that explanations of innovation and novelty were not ignored.  They were situated in disciplines such as comparative embryology, morphology, and paleontology (exemplified in the research of N.J. Berrill, D.D. Davis, and W.K. Gregory), which were overlooked because of a historiography emphasizing the relations between genetics and experimental embryology.  This identified the origin of Evo-devo tools (developmental genetics) but missed the source of its problem agenda.  The structure of developmental genetic explanations of innovations and novelties is compared and contrasted with those of other disciplinary approaches, past and present.  Applying the tool of conceptual clusters to these explanations reveals a unique form of conceptual change over the past five decades: a change in the causal and evidential concepts appealed to in explanations.  Specification of the criteria of explanatory adequacy for the problem agenda of innovation and novelty indicates that Evo-devo qua disciplinary synthesis requires more attention to the construction of integrated explanations from its constituent disciplines besides developmental genetics.  A model for explanations integrating multiple disciplinary contributions is provided.  The phylogenetic approach to philosophy of science utilized in this study is relevant to philosophical studies of other sciences and meets numerous criteria of adequacy for analyses of conceptual change.