Alumnus Alfred Blumstein, Ph.D. '60, Discusses his 40 Year Career as an OR-trained Criminologist

In a talk to the Boston Chapter of the Institute for Operations Research and the Management Sciences (INFORMS), Professor Alfred Blumstein described the role of Operations Research in the Criminal Justice System, which began with his service as director of a task force on science and technology supporting the 1965-67 presidential crime commission.

Alfred Blumstein completed his BS at Cornell in engineering physics and joined the Cornell Aeronautical Laboratory in Buffalo, earning an MA in statistics before returning to Cornell for PhD studies in ORIE. His research, under the guidance of Andrew W. Schultz, Jr. and graduate committee members Richard W. Conway, Lionel Weiss and Peter Ney, developed analytical models of airport capacity that could be used to investigate "the effect of various [design] parameters on capacity measures in order to indicate how best to improve the capacity," according to the resulting thesis.  

In 1965, after several years at the Institute for Defense Analyses, Blumstein joined the President's Commission on Law Enforcement and Administration of Justice, whose work led to a best-selling (and recently reissued) book, The Challenge of Crime in a Free Society. His leadership of the Commission's Task Force on Science and Technology began his career as a criminologist. (Unlike television's CSI forensic experts or Numb3rs math whiz Charlie Eppes, Blumstein focuses on macro aspects of the entire system, not the micro features of individual cases.)  In 1969 he joined the faculty of Carnegie Mellon University, becoming the J. Erik Jonsson University Professor of Urban Systems and Operations Research. From 1986 to 1993 he served as Dean of the H. John Heinz School of Public Policy at Carnegie Mellon.

In his talk to the Boston INFORMS group, Blumstein said of the 1960's task force that "we knew nothing of crime or the criminal justice system, but we brought a 'systems perspective,'" providing the Department of Justice (the only cabinet department without an assistant secretary for R&D) with a "window onto the then-remote world of science and technology." 

In the talk Blumstein showed a branched diagram of the flow of cases through the system -- starting with the crime and proceeding through police work, prosecution, the courts and the corrections system -- with the width of the lines proportional to estimates of the number of cases on each branch (the actual data was unavailable at the time). This diagram, published in The Challenge of Crime in a Free Society and more recently updated, may have represented the first such systems view of the criminal justice domain, and it influenced much work by Blumstein and others in the ensuing 40 years. The task force work also led to establishment of what is now the National Institute of Justice and the Bureau of Justice Statistics, the latter headed during the Clinton administration by Dr. Jan Chaiken (In his earlier career, as a Cornell mathematics professor, Chaiken taught mathematical analysis to at least two students influenced by Blumstein to pursue PhD degrees in ORIE). 

Blumstein discussed the use of simulation modeling in estimating costs, assessing the impact of policy changes, determining resource needs and understanding the impacts of demographic shifts. The early models led to some important policy studies, for example predictions of the burden on courts and prisons of increases in drug arrests as well as the analysis of initiatives in New York City to prosecute so-called quality-of-life crimes. More importantly "use of the model got people to think about the components as part of an interacting system," according to Blumstein.

However the early models did not account for feedback in the form of recidivism, treatment and reentry on behalf of individuals in the system. Adding feedback to the model led to the analysis of "criminal careers," according to Blumstein, and introduced an essential parameter that represented the frequency of criminal offenses by an individual. (This rate parameter of the "crime-committing stochastic process" was represented by λ, a notational convention that was "greek" to some criminologists, according to Blumstein). Nonetheless, estimating λ for various populations and at various ages has become a major focus of crime research.

Much earlier, Blumstein had discussed the implications of modeling criminal careers, particularly with respect to the rehabilitative aspects of incarceration, in a 1981 article in the Engineering: Cornell Quarterly, noting that "new analytical skills were needed to augment those of psychologists and sociologist, who had been in charge of the experimental evaluations of rehabilitation: experts in techniques such as econometrics and the analysis of stochastic processes began publishing research related to crime control."

The feedback model led to a number of policy insights described by Blumstein in his Boston talk, such as the notion that sentencing decisions should take into account the 'residual career length' of convicted law-breakers so as to avoid wasting prison space. 

Blumstein also presented data on incarceration rates ("the US is now the 'world champion,' having recently taken the lead from Russia") by crime type, and has argued since at least 1992 that incarceration of drug offenders is futile and even counter-productive because "resilient markets would recruit replacements" who may be "worse than the original offenders." He noted that the primary factors driving the growth in prison populations has not been increases in crime or in arrest rates, but rather increases in the proportion of arrests resulting in commitment to prison and in the length of sentences. Both of these are important policy choices under control of the criminal justice system.

Blumstein then discussed the unintended consequences of incarceration - an increase in drug offenders in prison, recruiting of young replacements by crack markets, the proliferation of guns to protect the markets and the tendency of young men to "resolve disputes by fighting" with much more lethal weaponry. In fact, Blumstein posited, the entire homicide rise from 1985 to 1993 can be explained by the increase in killings by young people with handguns, many in the drug trade.

Blumstein concluded his talk with an explanation of the drop in the crime rate after 1993 and an enumeration of factors potentially creating a new crime rise, including a drop in the economy and the availability of jobs for young people, the "diversion of police to terror and its alarms," new violent drug markets, the diffusion of guns among young people, a decline in state and local government revenue leading to a decline in social services, and a slowing of the growth of incarceration rates.  He warned that the model now predicts a rise in crime rates due to changes in several input factors.

Les Servi, Chair of the Boston INFORMS chapter that hosted the talk and a member of the national INFORMS Board of Directors, said that "the talk was fascinating to me on two levels: it was a chance to see cleverly designed studies which illuminate elements of the criminal justice system and impact policy..and to see how Professor Blumstein led the charge to create a science in an area where there was none before."

According to a 2007 article in the journal Operations Research, Blumstein considers himself a "missionary" to the criminal justice system, bringing Operations Research perspective to "those who have not yet adopted quantification, modeling, system perspectives and planning that characterize the hallmark of OR). He notes that "it has always been clear that [OR missionaries] can improve performance to some degree using the methodological skills that they bring to the table...[but] the new perspectives they bring...can be at least as valuable by raising fundamental questions about the operation of the systems they begin to investigate." In 1992, as President of the American Society of Criminology, Blumstein noted in his address that "the criminal justice system is behaving irrationally by any criterion," and urged major efforts to replace "fear and punitiveness" as policy drivers by "major efforts... to identify the nature of the problem, to assess alternative approaches, and to restore rationality to the policy process." More recently, he has testified about the irrationality of the 100 to 1 disparity in the sentencing guidelines for cocaine in "crack" versus powder form. 

In addition to his role as an eminent criminologist, Blumstein has served as President of the Operations Research Society of America (ORSA), The Institute of Management Sciences (TIMS), and the organization resulting from their merger, the Institute for Operations Research and the Management Sciences (INFORMS).    In 1998 he was elected to the National Academy of Engineering. He and his colleague, psychologist Terrie E. Moffitt, were awarded the 2007 Stockholm Prize in Criminology for their work on the analysis of criminal careers -- the first Americans to win this prestigious prize, which Blumstein jokes is a "deci-Nobel," since the monetary award they shared is a tenth of the Nobel prize amount.

Blumstein returned to Cornell in 2001 for his 50th undergraduate reunion, and has attended Cornell graduations of two of his daughters. In 2002, he participated in an external Program Review of ORIE, a process that is required of all academic units of the university.

Professor Saul I. Gass of the University of Maryland, a member of the Task Force that Blumstein headed and among other things a chronicler of the history of operations research, notes that "today, a mathematician helping police to solve crime ...seems ordinary. But in 1966, finding a scientist within the criminal justice system was rare. By some stroke of luck, Al Blumstein, a Ph.D. in operations research, was chosen to be the director of ...the Task Force. Al's systematic view of the interactions between the courts, police and corrections has proven to be a seminal and lasting contribution. This came about not by theoretical musing in the office, but by Al's scientific philosophy: learning and assimilating everything he could of the system... short of getting arrested, prosecuted, and tried."