AI helped me with this post.
Middle East Studies gets described constantly and counted rarely. Critics say the field trains students to work on Gaza, empire, and American power. Defenders say outside pressure is making honest scholarship harder. Both sides read the field from its controversies.
The public record says something else, and it is checkable. Departments publish dissertation titles. Some publish placement records. Job boards archive what universities were willing to pay for. Syllabi show which books carry the field from one generation to the next. A pilot of 153 dissertations and one department’s placement tables now exists, assembled entirely from pages anyone can open.
Two findings come out of it. One is solid enough to publish. The other is smaller than it looks at first sight.
The finding
Princeton’s Department of Near Eastern Studies publishes the first position and the current position of every PhD it has awarded since 2000, broken into five-year cohorts. Across the four completed cohorts, the way a graduate enters the profession has been replaced.
Of the 33 PhDs awarded from 2000 through 2006, 15 went straight into tenure-track positions and 7 took postdoctoral fellowships. Of the 20 awarded from 2007 through 2011, 6 went straight to tenure track and 8 to postdocs. Of the 23 from 2012 through 2016, 4 and 13. Of the 17 from 2017 through 2021, 1 and 12.
Direct entry fell from 45.5 percent to 5.9 percent. Postdoctoral entry rose from 21.2 percent to 70.6 percent. Both movements are monotone across every cohort. A Cochran-Armitage test gives z of -3.225 and p of 0.0013 for the decline, and z of 3.662 and p of 0.00025 for the rise. Comparing the earliest completed cohort against the latest, Fisher’s exact test gives p of 0.0045 and an odds ratio of 13.3.
A fifth cohort sits on the same table, running from 2022 through June 2026, and it moves the other way. Two of its 19 members entered tenure-track jobs, which is 10.5 percent. The cohort is open, its newest members have had months rather than years, and 2 of 19 against 1 of 17 is noise on those denominators. The trend tests above use the four completed cohorts and leave it out. Anyone checking the source table should see it there.
The doctorate qualifies the holder to begin a second apprenticeship of unknown length.
What the finding is not
Read the first-position table alone and it looks like an occupational collapse. Princeton also publishes current position, and that reading dies immediately. Of the 2000 through 2006 cohort, 25 of 33 now hold tenured or tenure-track jobs. Of 2007 through 2011, 16 of 20. Of 2012 through 2016, 16 of 23. Even of 2017 through 2021, whose members are five to nine years out, 10 of 17.
Many graduates arrive. They arrive later. First position and current position have to be reported together, because each alone tells a false story: one of a profession that has stopped hiring, one of a profession where nothing has changed.
What the tables cannot say is how long the delay runs. They are cohort aggregates. They hold no transition dates. One person may have spent a single year in a fellowship and another eight, and the published numbers cannot distinguish them. Any sentence about how many years the apprenticeship lasts is an invention until somebody builds the career histories.
The control, and what it did to the interpretation
The obvious question is whether this belongs to Near Eastern Studies or to the humanities. Princeton’s own History department supplies the nearest comparison, since it faces the same institution, the same job market, and the same access to prestigious fellowships.
History reports that among 2021 through 2025 graduates who went into education, 66.7 percent took postdocs and 7.4 percent went to tenure-track jobs. The department publishes no headcount, but the rounded percentages resolve cleanly to 54, 13, 6, 3, 3, and 2 out of 81. Among recent Near Eastern Studies graduates entering academic or educational first positions, 13 of 16 took postdocs and 2 of 16 took tenure-track jobs. Fisher’s exact test on the postdoc comparison gives p of 0.376. The intervals overlap heavily.
The comparison rules out a large difference between the two departments and is too weak to rule out a moderate one.
Two further things on the History page.
The first supports the delay reading rather than the collapse reading, from an independent department. Among History PhDs from 2008 through 2020, observed five or ten years out, 83.7 percent of those in education hold tenure-track positions and 1.1 percent remain postdocs. The apprenticeship resolves. At an elite department, it appears to resolve for most people.
The second is a warning about the comparison. History’s denominator counts only graduates who entered education, and education accounts for 78 percent of its PhDs. Princeton’s Near Eastern Studies denominator counts every graduate. The two departments publish different objects. History’s numbers come from web searches and emailed self-reports with no stated headcount and a measurement window that shifts by cohort. Near Eastern Studies publishes what reads as a departmental census with first position defined the same way for twenty-six years. A test that compares a census against a reconstruction from rounded percentages should not be reported as though it settled anything.
The national picture points the same direction. Humanities Indicators, drawing on the Survey of Earned Doctorates, reports that postdoctoral commitments among new humanities and arts PhDs rose from about 4 percent in 1990 to about 12 percent in 2020, with history above 18 percent. Princeton’s rates are far higher, and Princeton graduates have unusual access to fellowships. The comparison suggests without settling.
Why the study still belongs in this field
If postdocization runs across elite humanities, then the question it raises runs across elite humanities too. A scholar on a one-year appointment depends on recommendations, renewals, invitations, and the judgment of future search committees. A tenured professor can anger a donor or a colleague at lower cost. The formal rules permit dissent in both cases. The material structure prices it differently.
Middle East scholars report high rates of professional self-censorship. In late 2023, 82 percent of United States respondents said they self-censor when speaking professionally about Israel and Palestine, and among assistant professors the figure reached 98 percent. In the round fielded around the start of the 2026 Iran war, 77 percent reported self-censoring, and self-censorship on Iran doubled from 17 percent to 33 percent overall and 40 percent in the United States within weeks.
Those numbers are read as evidence about ideology. One camp sees fear of donors, legislators, and advocacy groups. The other sees fear of colleagues enforcing a consensus. Both may be right. Neither has been separated from a third possibility, which is that people who cannot count on employment past next June behave differently from people who can.
Most fields poll their members about nothing. Middle East Studies has a running survey of professional speech, conducted by people with the standing to add questions to it, and now the beginnings of an employment record built from public pages. The field is the natural laboratory for the question precisely because it is unusual in what it measures about itself, and the answer would apply well beyond it.
That study can embarrass everyone. It might show external pressure dominates. It might show internal peer pressure dominates. It might show the labor structure amplifies both.
The dissertation finding, and how far it reaches
Princeton and Chicago together produced 123 dissertations across the sampled years. Coded against 1948, 101 fall before, 3 after, 5 straddle, and 14 cannot be placed from the title. Among clear cases, 2.9 percent are post-1948. Anyone who believes these departments turn out scholars of contemporary Palestine has to explain the Ottoman postal service, Ugaritic pronouns, Hittite ceramics, and the marginalia of a sixteenth-century Cairene scholar.
The obvious objection is that Chicago trains philologists, archaeologists, and Egyptologists, so the sample guarantees the answer. Strip out the ancient, biblical, and Egyptological work and 69 clear cases remain, of which 3 are post-1948. The share moves from 2.9 percent to 4.3 percent. The objection fails.
Adding Columbia’s interdisciplinary department was meant to test whether interdisciplinary training produces contemporary work. Among Columbia’s recent Middle East and North Africa dissertations that classify cleanly, 4 of 14 are post-1948, or 28.6 percent. Columbia differs from the two traditional departments, and the difference is unlikely to be noise, with Fisher’s exact p of 0.0138.
Where Columbia sits remains unknown. The Wilson interval on 4 of 14 runs from 11.7 percent to 54.6 percent. It contains 50 percent. The data can say Columbia is more contemporary. The data cannot say whether the field contains two separated clocks or one gradual slope, and any draft that picked between them was overreading its own numbers.
What the sources taught, which may outlast the findings
Three departments published placement information and produced three different objects. Princeton distinguishes first position from current position and defines both. Chicago heads a column Job Placement without saying which it means. Columbia publishes current position only. Pooling them produces nonsense, and the temptation to pool them is strong because they look alike in a spreadsheet.
Columbia’s roster lists 8 graduates for 2020, 1 for 2021, 10 for 2022, 7 for 2023, and 4 for 2024, and the page’s own metadata gives a last modification date of August 14, 2024. A department producing eight to ten doctorates in neighboring years did not produce one in 2021. The roster is maintained rather than complete, which means it cannot serve as a production series, and if the missing names are the people who left the profession, the placement column reads high.
Columbia publishes no dissertation titles. Six had to be recovered from elsewhere, and three were wrong. In every case a later book or fellowship project title had displaced the deposited dissertation title. Scholars with books and fellowships leave a bigger public trace, so reconstruction from visibility oversamples the successful, and the error correlates with the outcome anyone would want to measure.
What needs doing, in order
The bottleneck has moved from analysis to collection, and further interpretation of the existing numbers will not pay.
The first job is person-level career histories, and it is the one that converts a suggestive finding into a measurable one. Princeton’s Near Eastern Studies department publishes the names and dissertation years of every PhD since 1927. For the 112 graduates since 2000, a CV, a faculty page, or a professional profile gives the doctorate year, each position with its dates, and the year of first tenure-line appointment. That yields median time to a permanent job by cohort, and the share still outside a tenure line at one, three, five, and eight years, with proper handling of people whose careers are still running. Roughly 112 people, perhaps twenty hours, and it answers the question everything else has been forced to leave open. Do the same for Princeton History, whose doctoral degrees awarded page supplies the names, and the control becomes a real one rather than a comparison of unlike published summaries.
The second job is the dissertation hypothesis, and department pages are the wrong tool for it. The strong version predicts the largest contemporary share inside political science, anthropology, and history departments, which do not publish Middle East title lists at all. A ProQuest or repository query returns institution, year, and title for several hundred such dissertations in one operation, without depending on any department’s willingness to maintain a page.
The third job is job advertisements, still untouched and still the only measure of demand rather than supply. Archived captures of the H-Net job guide and the MESA employment page reach back past 2016. Retrieval is the entire difficulty, and retrieval is solved.
The fourth job is the survey, and it needs a collaborator rather than a scraper. A future Barometer round that asked about tenure status, contract length, renewability, years since the doctorate, years spent in temporary appointments, visa dependence, institution type, and reliance on regional access would let anyone test whether employment vulnerability predicts silence once political position and subfield are held constant. The people who run that survey have the standing to add the questions. Nobody else can.
Syllabi come last, and possibly never. The access cost is high and the return is the lowest of the five.
What would kill each claim
The postdocization finding dies if the person-level histories show that the time from doctorate to permanent job has stayed flat, meaning the postdoc was always there and only the labeling changed. It also weakens sharply if Princeton History’s cohort-by-cohort curve, once reconstructed, turns out to be steeper than Near Eastern Studies.
The precarity hypothesis dies if self-censorship shows no relationship to contract security once beliefs and subfield are controlled. That would be a real result and would strengthen both political explanations at once.
The two-clock hypothesis dies if disciplinary departments turn out to produce pre-1948 work at rates near Princeton’s and Chicago’s. It survives in its strong form only if their contemporary share clears 50 percent with an interval that excludes the traditional departments.
Why any of this is public business
The public pays for the universities, the fellowships, the libraries, and much of the student support. Students organize six or seven years of their lives around a career, and now a further stretch of unknown length after that. Departments admit doctoral candidates without publishing, in most cases, any honest account of what happens next. Princeton’s Near Eastern Studies department is unusual in publishing enough to be analyzed, and it comes out of the exercise looking honest.
Arguments about this field almost always start from a political conclusion and gather examples. The counting runs the other way. It asks what departments produce, what universities advertise, where graduates go, how long they stay temporary, and which books get assigned. The answers so far have embarrassed the loudest claim on offer, which was that traditional departments have been captured by present-day politics. They have also failed to support the tidy counter-story about area-studies funding.
The numbers are public. The methods are ordinary. The employment and production record of this field has never been assembled, and assembling it takes weeks rather than grants.
Notes
Version 0.5, posted July 26, 2026. Cite this version and date, since the files below will change as the collection grows. Every count was computed from the released files rather than typed by hand.
Corrections are welcome and will be published here with attribution. The coding of individual dissertations, the Columbia roster gaps, and the three recovered titles marked provisional are the places where somebody inside these departments can settle a question in a sentence that would take me a week.
Data
middle_east_studies_incentives_v04
mes_princeton_internal_controls_v04
Sources
Princeton Near Eastern Studies first positions
Princeton Near Eastern Studies current positions
Princeton Near Eastern Studies past dissertation titles
Princeton History career outcomes
Princeton History doctoral degrees awarded
Chicago Middle Eastern Studies alumni placement
Columbia GSAS dissertation notices
