Skip to main content


ICC 650
Box 571014

37th & O St, N.W.
Washington, D.C. 20057

maps & directions

Phone: (202) 687.6400



The Interdisciplinarity Movement

Posted on

My current reading about research collaborations presents the puzzle of why a movement for interdisciplinarity is occurring at this time. Certainly, if one examines the current set of unsolved world problems (e.g., income inequality, environmental impacts on society, terrorism), they all seem to clearly present themselves as complex systems of interrelated features. But, how could anyone ever have thought that they could be “solved” by interceding on only one of the features? Did not scholars always see these problems as requiring solutions-based knowledge from multiple domains? Were academics unmotivated to tackle such problems?

I have multiple hypotheses for why we see such current support for interdisciplinary work. First, the call for interdisciplinarity might be the unintended result of reductionist progress of all the disciplines. It’s pretty easy to document that academic fields tend to multiply, as knowledge evolves. We have more undergraduate majors and more graduate programs than we have had in prior decades/centuries. Human knowledge just doesn’t expand; it spawns relatively cohesive subfields, like the mitosis of cells. Subfields provide the environment for deeper and deeper study. Some of the divisions, however, take on more applied problems (e.g., computer science emerging from electrical engineering, finance programs emerging out of business schools and economics departments, environmental biology). Sometimes these sub-disciplines are better thought of as “inter-disciplines.” They then stimulate, in their journals and texts, a platform of thought compatible with interdisciplinary research. Once an openness to interdisciplinarity is bred, it tends to shape lifetime agendas.

A second hypothesis comes from the organization of funding agencies of central governments and international organizations throughout the world. One could make the argument that social and political oversight of these institutions has demanded greater societal benefit. Governments need to identify how academic research benefits the taxpayers. It’s easier, the argument goes, to justify research activities that solve well-documented societal problems. Application of knowledge thereby enjoys preference over the generation of basic knowledge. Hence, we see a disproportionate number of new initiatives in science funding agencies aimed at solving well-defined problems (e.g., search for a universal influenza vaccine). These problems demand interdisciplinary efforts.

A third hypothesis concerns a growing commonality of research methods. When two fields use the same research tools, it’s easier to implement collaboration across fields. The invention of functional magnetic resonance imaging (fMRI) devices has led to collaboration among biomedical researchers, cognitive and social psychologists, and even behavioral economists in ways that would be unimaginable prior to the invention. The growing use of social media data in the social sciences is another example, with data scientists, macroeconomists, social psychologists, political scientists, and communications science researchers using Twitter, blogs, and other data. One of the great problems of effective collaboration across disciplines is the lack of common language. Using the same research tools eliminates some of this problem.

A fourth hypothesis is the call on universities to become more relevant to the societies in which they are placed. (This is similar to pressures on research funding agencies.) The invention of land grant universities in the US in the 1860s created institutions with explicit missions of serving their communities. Jesuit institutions have explicit missions of service; some other institutions share this. But on the whole, as public opinion in the US reflects, universities are viewed by many as irrelevant to their day-to-day lives and ineffective in improving their lives. There are more calls by social and political leaders for university engagement in the critical issues of the day. Indeed, Georgetown’s efforts at interdisciplinarity are often motivated by our desire to improve the lives of others. Societal impact as an explicit goal breeds interdisciplinarity.

The movement toward interdisciplinarity is likely a complicated blend of all these features of modern life. Certainly, the “why” of interdisciplinarity in universities is equally important as the “what” of interdisciplinarity. Strong disciplinary knowledge is needed for interdisciplinary work. Sometimes basic findings in a single discipline find their interdisciplinary application decades later. Leaving one’s disciplinary home needs a lasting reason. When the “why” is a passion for solving an important problem affecting the world, collaborating across disciplines is most easily sustained. I like that about Georgetown.

Remember Those Off the Grid

Posted on

There was a recent Pew Research Center study that caught my eye. It reports that fully 11% of US adults do not use the Internet.

These days, social media communication platforms appear to play a more and more important role in politics, commerce, and friends’ network formation. But the study is a useful reminder that we may be misled by information solely based on these platforms when we’re forming impressions about the entire US adult population. In fact, for those of us who are nearly constantly using the Internet for our day-to-day life activities, it’s difficult to imagine that not everyone lives like we do. It’s easy to assume that reports about behaviors on the Internet represent everybody in the country.

Now, it’s important to note that the percentage of adults not using the Internet has dramatically declined over time, from 48% in 2000 to the 11% this year, using roughly the same survey method to generate the statistics. However, an increasing number of the statistics that the news media present to us each day, which we all use to form impressions of what the US public thinks about topics, are based on those with Internet connections.

So, what do we know about those who do not use the Internet? The study shows that they disproportionately are rural residents (the rural rate of nonuse is 22% versus urban rates of 8%). Part of this result is no doubt associated with the inadequate penetration of high speed Internet and spottier cellular services in rural areas. Indeed, much of attraction of the Internet comes with fast response and ubiquitous availability throughout the day.

Education attainment is another correlate of nonuse of the Internet, with adults having less than a high school education showing the highest rate of nonuse (35%). Since lower education attainment is also correlated with lower income, the study also finds that the lowest income group reported (those earning less than $30,000 per year) are nonusers at a 19% rate. With these two attributes combined, one is reminded that Internet use costs money out of one’s personal budget each month, and Internet connection also has greater value to those desiring ubiquitous, continuous access to information sources.

Finally, age is correlated with nonuse. Thirty-four percent of the highest age group reported, 65 years old and older does not use the Internet. As this group passes out of the population, there will be dramatic changes in the rate of nonuse (the next highest age group, 50-64 year olds, have only 13% nonusers). It’s easy to speculate that in 10 years or so, the 11% overall nonusers will dramatically decrease.

The study shows a fascinating lack of correlation between Internet use and race/ethnicity, especially when one considers that income differences among such groups.

So, what does all this mean? We need to be careful to remember that, for the time being, conclusions reached about US adults based on Internet behaviors miss 1/10 of the population. This 11% is older, poorer, less educated, and more rural than the rest of the population. When an Internet study describes behaviors that we would expect to vary across age, urbanicity, income, and education groups, we should be cautious about the conclusion. So, the next time we see a report on what is circulating on Facebook, what popular searches are for Google, and what’s trending on Twitter, remember that 11% of the population is completely divorced from such behaviors. It’s important to read the details of how the data were collected before accepting the headline of the story.

Conflicts between the Creator and the Audience

Posted on

I was recently reading a discussion about artists’ reactions to commentary on their work. While many artists view their work as occupying a place of its own, removed from the reach of discourse, their audience sometimes feels obliged to analyze it, dissect it, categorize its various attributes, and induce from those attributes the central meanings and intents of the creator. I recall artists saying, “If that’s what you see in it, that’s ok with me.” Other artists find commentary dissecting their work as mainly irrelevant to what they believe they are doing in life. Still others are offended. Clearly, similar reactions occur, but perhaps at a different rate, in the literary arts, where words rather than materials are the medium.

However, it does seem when a piece generates intense and long-lasting attention by the audience, it signals an important contribution. Work that is discussed and dissected by hundreds of audience members over years generally is viewed as valuable.

I wondered whether there are similar distinctions between the creator and the audience in the social sciences and natural sciences. These domains of creation are governed by more well-defined paradigms, greater constraints on forms of presentation, and heavily governed norms of conclusions. The act of creation required for the product may be more linked explicitly to past work; its uniqueness has less value.

Further, while some of the power of art is its ability to generate different reactions across audience members, much of the focus of scientific papers is guiding the reader to a single reaction. The desired reaction is a certainty that the conclusion of the writer is correct.

But research papers do sometimes generate diverse reactions. This generally occurs in one of three ways. First, readers may question the conclusions as being unjustified by the presented evidence. Such reactions often occur during the peer-review process that precedes the dissemination of a research product. If such skepticism is displayed by several reviewers, the product is not disseminated by the publication organ.

Second, the readers may question the integrity of the data generation process. If this occurs during peer review, it may similarly yield a rejection by the journal. If an article appears that generates such reaction, it can sometimes stimulate another researcher to mount a rebuttal to the original paper.

Third, the readers may question whether the required assumptions underlying the research findings are plausible. When this occurs, other work is also stimulated, but this generally is work that portends important developments might possibly be occurring. Many fields wobble back and forth between periods of relative stability and periods of controversy and ferment. In the times of stability, there is consensus on the important questions and the appropriate methods to address the questions. But all approaches to inquiry rest on certain assumptions. When a field begins to questions those basic, well-accepted assumptions, look out for controversy! Further, when a paper generates intense and long-lasting discussions challenging long-held assumptions, it is often viewed as an important contribution.

So, compared to artists facing critics deconstructing their work into constituent components, the world of social and natural sciences seems inevitably different. Acts of creation explicitly build on one another over time dialectically. Critical dissection is desired and mandatory. That is how the goal of the creator (the conclusion he or she reached) is widely acknowledged as having been achieved. Where the two domains of creation come back together, however, is the fact that intense controversy and long-lasting discussions about a piece of work generally means something important has been created.

Three Levels of Understanding

Posted on

I taught the same course for about a ten-year period, sometimes twice a year. My recent reading about “threshold concepts” in different disciplines has made me rethink my own evolution in understanding the material in the course I taught. I now perceive three stages of my learning.

First, as a graduate student, I conquered all the major concepts and their interrelationships. This meant I could answer any question that an instructor could pose to me; I could “solve all the problems at the end of chapters;” I could read the research literature in the field. In retrospect, this was the level of understanding that only made me a good student in an advanced course in the field. I now realize how superficial such understanding really was.

Second, post-PhD, after teaching the course a few times, I realized a different level of understanding. Upon reading just a few words of a question, I could classify it into its proper context among all the concepts in the field. I could mix and match concepts in creating new questions, to test students’ ability to synthesize them. Part of this understanding arose from lecturing on the different components of the field and inventing progressively complicated constructions of pieces of knowledge as the course proceeded. Concepts built upon concepts, creating more possible combinations. Teaching led me to relearning the material, but in a new way.

As I assembled class examinations over the years, I forced myself to construct different versions of questions, each of which tested the same subdomains. The questions were different enough from one another that clever students who acquired prior examinations could not profit from knowing prior exam questions. Indeed, in later years of the course, I would give out to students many of the questions of prior examinations in order to help them study. My hope was that their exposure to hundreds of questions related to the field could indeed give them this same second level of understanding. They too would be able to quickly recognize what combination of knowledge a given question was seeking.

The third level of understanding came later in my teaching of the course. After grading thousands of examinations, I was exposed to all the possible wrong answers to a given type of question. I had seen every possible mistake. Further, in talking with students after exams, I learned what errors in logic led to the wrong answers. When I was at my best helping a student, I was able to identify the problem in their logic, go back to the material that troubled them and correct it. These episodes were often transformative for the student; everything became clear. A roadblock had been removed. In some sense, I felt that was my best teaching. Many of the causes of errors stemmed from their lack of understanding at the first level mentioned above. They never really “got” one of the early concepts in the course, leading to confusion with diagnosing how best to approach a given question much later in the course.

This problem is quite close to the notion of “threshold concepts” made popular by David Pace and his colleagues. They make the argument that there are key concepts in every field that are gateways to more sophisticated understanding. The more structured a given field, the more likely that there are key central notions that are building blocks to the entire field. One of the surprising findings of the research is that some sophisticated experts in a field cannot easily identify these threshold concepts. They find it difficult to remember the time when they learned the material for the first time. They look upon the field with completely different perspectives than that of a novice. Hence, the relative difficulty of key concepts to comprehend upon first exposure is something they have forgotten (or maybe never perceived).

One of the ongoing challenges of teaching, it seems, is remembering the perspective of the learner, not the expert in the field. The learner is constructing the conceptual structure, brick by brick. The expert sees from above the entire, integrated structure, in all its beauty.

Coming Closer to the Census

Posted on

As we approach 2020, my past life as a Census Bureau director begins to intrude, with calls from journalists about my thoughts on the funding, design, and politics surrounding the next census. Each decade about this time (the year ending in “8”), funding issues come to the fore, concerns over the latest census technological innovation arise, and a re-education takes place for political elites about the independence of the operations of the census. This decade is no different. This is the year, under the current laws and regulations, that the final form of the census questions is selected and verified by Congress. This is the year that funding must ramp up for the large-scale field data collection operations to contact households that don’t themselves answer the call when first made. This is the year that, like it or not, key risks of the next census start to come clear.

While there are complicated statistical and data processing issues involved in any census, my memories of the experience arise from completely different aspects. One of the duties of a director during the census is to promote civic participation in the census, encouraging each household to provide information requested by the census. The director does a lot of traveling, focusing time and attention on communities where prior research has shown to be reluctant to respond to the census.

I spent time on the Texas-Mexico border. I visited several “colonias,” which are collections of self-made housing, often lacking electricity and water services. Many are placed on unused fields of ranchers and farmers, who rent small plots on which the structures are built. The extent of undocumented persons is high. Fear of the Federal Government is widespread. Many residents have fled countries in which the government could not be trusted. In this environment, we attempted to convey the message that all persons are part of the US census, whether they are citizens or not, whether they are documented or not. I tried to assure them of the strong legal protections of data confidentiality. I found families with admirable resilience, communities with very strong bonds, and households that shared among themselves in admirable ways. Indeed, they had come to the country for the same reasons that many, many others came to the US: the hope that the next generation would have lives that were safer and more financially secure than the current generation.

I spent time with Muslim-American and Arab-American groups in multiple cities. I will never forget a dinner with such a group. As the dinner progressed at a table of 10 or so persons, and we became more comfortable with one another, the father of a large family asked what the question about “race” should mean for them. Did the census seek a self-report of skin-color? Was it something else? It reinforced the fact that the social constructions of the American experience were not ubiquitous to all cultures now in the country.

I spent time with groups connected with various islands of Micronesia, each with strong cultural bonds both within their US communities but also with their islands of origin. They sought recognition of their numbers within the larger fabric of the country. They were proud of their hyphenated American community and looked to the census to make known to the society how their numbers had increased over the previous ten-year period. They too sought the opportunities for their children in the US and were full believers in the American Dream.

These groups on the whole were not rich in money, but they were wealthy in the bonds of family, friends, and community. It made me proud of how the country is enriched by their contribution to the social fabric.

A Survey that Really Matters: the 2018 COACHE Survey of Faculty

Posted on

Since 2013, every two years, we mount a survey of full-time faculty on the main campus and the Qatar campus, using a web-based questionnaire organized by the Harvard School of Education. In doing this, we join Johns Hopkins University, Duke University, Tulane University, University of Rochester, University of Arizona, and University of Virginia, among others.

The survey is called the Collaborative on Academic Careers in Higher Education (COACHE) Survey. The areas covered by the survey include the research environment and institutional support; teaching load and student quality; nature and distribution of service responsibilities; facilities and work resources; personal and family policies; health and retirement benefits; interdisciplinary work and collaboration; mentoring; tenure and promotion opportunities, clarity and reasonableness; recruitment and retention, and institutional governance and leadership.

I view this as a bit of a report card on the provost’s office. It has told us which aspects of faculty life are getting better and which are not. When concerns of faculty were evident in the statistical results, we assembled faculty task forces to ameliorate the problems. We mounted focus groups to get richer, qualitative information about important domains. We designed and implemented improvements in areas where we saw major problems. While there is always more work to do, we’ve made some progress on the clarity of tenure and promotion criteria, support for interdisciplinary scholarship, and mentoring for associate professors.

We’ve found the systematic measurement very useful. Use of the Harvard team to conduct the survey allowed us to assure faculty that administrators, like the provost, would never be able to associate answers with individual faculty members. Only statistical summaries of data are available to the provost’s office, deans, or unit heads. We sought transparency by giving presentations on the results of the survey to faculty throughout campus.

We want to continue using feedback directly from faculty, based on these measures, to hold ourselves accountable for the responsibilities of the provost’s office. We want to see whether things are getting better, staying the same, or getting worse.
Surveys become most useful with widespread participation by the faculty.

Please take a few moments to complete the web survey if you receive a request to do so. You can be assured that the many of your colleagues are participating; the results will be used, not just filed away. This is your chance to have a voice in the future of Georgetown.

The 2018 Provost’s Distinguished Associate Professors

Posted on

Each year we start the process of naming Provost’s Distinguished Associate Professors by having departments and similar units nominate deserving colleagues. A team of University and Chaired Professors judge the various nominations. They forward selections to the Provost. These designations are term-limited with a duration of five years, maximum. (When a Provost’s Distinguished Associate Professor is promoted to full professor status, the term would also be completed.)

For the third consecutive year, we are recognizing Associate Professors who are performing at such a high level that their colleagues have identified them for more visible praise. As indicated below, their work exemplifies what makes Georgetown strong – faculty thoroughly engaged in pushing the envelope of knowledge in their field, and transmitting their passion for such work to their students and the general public.

Caetlin Benson-Allott is an Associate Professor of English and core faculty member of the Film and Media Studies and American Studies programs. She joined the Georgetown community in 2010. Her research has focused on US film history since 1968, film and media theory, exhibition and new media technology, and gender studies. Dr. Benson-Allott is the author of Remote Control (Bloomsburg Press, 2015) and Killer Tapes and Shattered Screens: Video Spectatorship from VHS to File Sharing (University of California Press, 2013). In 2017, Dr. Benson-Allott was elected the new Editor of Cinema Journal, the Society for Cinema and Media Studies’ scholarly publication of record, a position she will hold through December 2022. Cinema Journal is the second-oldest scholarly film journal in the US and is the most esteemed film and media studies journal in the world.

Laurent Bouton is an Associate Professor of Economics. He obtained his Ph.D. from the European Center for Advanced Research in Economics and Statistics (ECARES) at the Université libre de Bruxelles in 2009, and joined Georgetown in 2013. Dr. Bouton’s research focuses on the incentives of voters under various electoral systems, namely the way in which they use information in these systems and how their use of information determines electoral outcomes. He has three publications in the top five general interest journals in economics, including: American Economic Review, considered by many as the flagship journal of the economics profession; Journal of Political Economy; and Econometrica, which is widely considered to be the top technical general interest journal. In addition to his strong publication record, in 2017 Dr. Bouton received a European Research Council (ERC) Starting Grant totaling 1.5 million euros. Dr. Bouton is currently a FNRS Research Associate at the Université libre de Bruxelles and a Research Affiliate of the Centre for Economic Policy Research. Additionally, he has served as a Faculty Research Fellow of the National Bureau of Economic Research since 2014.

Marcia Chatelain is an Associate Professor of History and African American Studies. Although her research specializes in African American children, race in America, and social movements, she has been recognized for her impact in issues outside her area of focus. Dr. Chatelain has been the recipient of a NEH grant and New America Foundation National Fellowship. She was a member of the Georgetown’s Working Group on Slavery, Memory, and Reconciliation, and her associated work with the group continues today. In 2016, Dr. Chatelain was named a “Top Influencer in Higher Education” by The Chronicle of Higher Education. She is the author of South Side Girls: Growing up in the Great Migration (Duke University Press, 2015). Dr. Chatelain joined Georgetown in 2011.

Bryce Huebner is an Associate Professor of Philosophy, a department in which he has served since 2009. His research has been highly interdisciplinary, including the study of interrelated issues in moral psychology and metaphysics of the mind; the cognitive strategies people employ in making moral judgments; and the ways in which neuroscientific and computational research on learning and motivation might be able to fund a more plausible account of moral cognition. In addition to his monograph, Macrocognition: Distributed Minds and Collective Intentionality (Oxford University Press, 2013), Dr. Huebner has authored 33 regular articles, seven shorter articles, and nine book reviews.

Please join me in congratulating these wonderful Georgetown colleagues for their accomplishments and their contributions to our university.

Facts, Evidence, and Changes of Perspectives

Posted on

In my ongoing attempts to read across the disciplines – reading sideways, in some sense – there are some consistencies in approaches across many fields that I had never observed before. Indeed, in my experience, some scholars in these fields are not particularly aware of these similarities. I’ve written whether facts are discovered or constructed. This post is about what we might mean by “facts” and “evidence” and how insights are dependent upon perspectives taken on a given issue. I think there are two ubiquitous practices of many disciplines that are deserving of note: 1) a use of concentrated observation, 2) the valuing of change of perspective for insight.

I was a member of the Commission on Evidence-Based Policymaking, which perforce needed to attend to differences between data and evidence. My own judgments after experiencing scores of discussions on these matters is that the difference between the two is important. Daniel Patrick Moynihan is credited with noting that, “You are entitled to your own opinions. You are not entitled to your own facts.” In some sense, however, formal evaluation studies entail the assembly of facts in a way that they inform a judgment about whether a program is functioning as intended. Facts (or data) are the precursors of evidence. Evidence is the coherent assembly of facts to permit a valid conclusion. In some sense, “evidence” requires an assembly of some facts, a discarding of some facts, and a synthesis of the chosen facts into a coherent insight. So, we could add to the Moynihan note: “Effective evidence demands wise selection of the available facts.”

This is relevant to disciplinary commonalities and differences. The first common practice appears to be an intense observational focus, the process of collecting and assembling facts. In the sciences that rely on deductions from well-found prior findings, the observation may examine an implication that had never been examined. Alternatively, it may be the focus on an untested assumption underpinning a given finding. In disciplines whose medium is text or images, the focus may be on a feature of the entity that was ignored in prior interpretations of the text or image. In disciplines examining human behavior, the focus may prompt measurement of thoughts or behaviors that offer new conceptual frameworks for explaining the behavior. Scholars succeed when such deep, focused attention is paid to the selective assembly of facts. When the assembly leads to a preferred level of coherence relative to the past work (e.g., a new theory, a new interpretation, a new discovery), an advance in the field is achieved.

The second common practice valued by scholars in many fields seems to be the ability to change perspective on a given issue. Mathematicians often note that breakthroughs came to them by entering the unsolved problem in a completely different way. Social scientists note that changing perspectives from a disinterested observer of the group to a member of the group can yield unrivaled insights. Those creating visual images are the masters of the importance of changing physical perspectives for the impact of a piece of work. Those intensely studying text often move closer and further away from ingredients of the text to unlock new interpretations. All of these seem to be examples of the scholar believing that new insights can come from changing the perspective taken on the issue at hand.

In my own education as a researcher, I can’t remember many formal discussions of these two practices. I do remember, however, adopting both, probably by mimicking the behavior of my mentors. It would be interesting to know whether collectively we are now better at educating young scholars in the value of these skills.

Discovering and/or Constructing Fact

Posted on

This is not a post about facts and alternative facts. It’s a post about how some fields determine what they currently believe to be true and what is too tenuous to believe.

I’m reading an older book, Laboratory Life: The Construction of Scientific Facts, written by a nonscientist observer in a neuroendocrinology laboratory. The observer attempts to glean how knowledge is built and assessed in the unit.

It’s always interesting to look at a process through the eyes of someone external to it. The value of such studies is that basic assumptions of a field, especially those that form the very basis of its foundations, are sometimes difficult for participants to articulate. One removed outside the milieu tries to discern these assumptions. When the observer can then seek verification of their conclusions, real insight is possible.

The book described how the group, in some sense, was in continuous debate, attempting to separate fact from artifact. Indeed, at the start of a new program, all possible facts might seem equally plausible. The consequence of research allows the identification of what is fact and what is artifact. Progress in the laboratory consisted of discarding possible facts, as quickly as possible. This is very similar to the notion of identifying the signal in noisy information or the notion of extracting order out of disorder. The observer concluded that resolutions of those debates on a given issue was a signal of progress in the inquiry. The classification as fact or artifact became part of their agreed-upon knowledge state, and onward the scientists went to studying the next issue.

The observer saw the scientists in the laboratory to be singularly focused on their impact on the field. They were continuously reading the results of peer/competing laboratories, and they were adapting their own agenda to be complementary to those. One principal means of having impact on the field was to produce scientific articles, peer-reviewed by those in the field. The achievement of the peer-reviewed article supported their belief that the finding in the paper was novel enough to merit a separate notification to the field.

Impact, however, was not just the publication but whether the findings reported in the article affected other people’s work. Citations to the papers were used as an indicator of this. But the citations tended to have multiple forms. Sometimes citations were made within articles that contested the original finding. Sometimes citations to the work were related to new developments building on top of the original finding. The latter, of course, were viewed as evidence of more positive impact.

Over time, if the percentage of work building on the original finding dominated the new literature, the scientists took that as evidence that their finding had lasting value. When a citation was found from a textbook providing an overview of the field, then the finding had, in some sense, become a fact. That is, the field was including the finding in teaching the key knowledge of the field. Further evidence of impact arose from the findings of the laboratory being used in clinical or pharmaceutical industry developments.

The subtitle of the book, noting the construction of facts versus the discovery of facts, is an interesting choice of words. I’m not sure all the scientists in the laboratory would view their work with that perspective. All fields in some sense are involved in such construction. Some fields are deeply self-aware of their role in the construction. Other fields hold fast to the notion that they are objective viewers of a fixed external reality. Their new facts are viewed to arise from discovery, not their own construction.

Hermeneutics and Bayesian Statistics

Posted on

I find myself reading across disciplines these days, with special attention to differences in the logical structure and scholarly criteria for determining excellent work in the fields.
From time to time in this learning, I find odd correspondences across two disciplines, sometimes two fields that tend to have little interaction. I hit one last week.

 The word, “hermeneutics,” a iterative process of discerning meaning in a text, is as old as biblical scholarship.  At this point the word applies to a whole sub field of philosophy, but also has more specific meanings. Its use in that sense involves a sequence of interpretations of a text – a “guess,” as it were – followed by application of the interpretation to a more detailed review of parts of the text, other texts of the same author, or external phenomena related to the text.  In short, is the initial interpretation supported by repeated, more specific examinations?  If not, the initial interpretation is altered and the process begins again, until at some point, there appears to be no important change in the then-achieved interpretation.  (There are some obvious caveats here.  One is that we must assume the text itself exhibits some coherence.)
A key attraction of revealing such a conceptual structure on interpreting the meaning of words is that it can help direct new students to their behavior as they learn to examine new material.
Bayesian statistics is founded on an important theorem, mathematical in its logic, that permits an integration of what is known prior to an analysis of new data into the analysis of those new data themselves.  Instead of basing our conclusions about a phenomenon (e.g., what portion of a patient population benefits from a specific drug) only on a given single set of data, our conclusions will ask the question of how the new findings alter the prior conclusions, based on other sources of data on the same phenomenon.  (There are some obvious caveats here, too.  One is that we must assume that there is no difference between the conditions generating the prior data and the conditions of the current data collection.)
So what?
First, it’s fascinating to me to learn of two, relatively independent fields, inventing methods that resemble one another. Second, one wonders about the counterfactual – what would have happened if the two fields had been collaborating earlier?  For example, the application of hermeneutics, in some sense, seems quite adaptive to new information.  Indeed, some treat the structure of hermeneutics as a circle of interpretation/reinterpretation that never ends.  New observations can be entered into older completed interpretations, yielding a new state of interpretation. 
Bayesian statistical approaches are designed as a two step process, an integration of the new observation “on top of” everything else we know that yielded our beliefs prior to the new data.  Of course, repeated application of Bayesian estimation to repeated new data collections creates on going updating process that closely resembles the hermeneutics circle.  But its original focus was the two step process.  

It looks like some developments in machine learning, which can produce constantly updated predictions of the future state based on new data, are using formal Bayesian methods. This resembles more fully the continuous revision of the hermeneutics circle.

So maybe the two conceptual structures are increasingly resembling each other.  I’d love to see a dialogue between these fields to see if any new thoughts would arise if they understood each other more fully.

Office of the ProvostBox 571014 650 ICC37th and O Streets, N.W., Washington D.C. 20057Phone: (202) 687.6400Fax: (202)

Connect with us via: