Open Patrologia Graeca 1.0

August 8, 2015
Comments to

Federico Boschetti, CNR, Pisa
Gregory Crane, Leipzig/Tufts
Matt Munson, Leipzig/Tufts
Bruce Robertson, Mount Allison
Nick White, Durham (UK) (and Tufts during 2014)

A first stab at producing OCR-generated Greek and Latin for the complete Patrologia Graeca (PG) is now available on GitHub at This release provides raw textual data that will be of service to those with programming expertise and to developers with an interest in Ancient Greek and Latin. The Patrologia Graeca has as much as 50 million words of Ancient Greek produced over more than 1,000 years, along with an even larger amount of scholarship and accompanying translations in Latin.

Matt Munson started a new organization for this data because it is simply too large to put into
the existing OGL organization. Each volume can contain 250MB or more of .txt and .hocr files, so it is impossible to put everything in one repository or even several dozen repositories. So he decided to create a new organization where all the OCR results for each volume would be contained within its own repository. This will also allow us to add more OCR data as necessary (e.g., from Bruce Robertson, of Mt. Allison University, or from nidaba, our own OCR pipeline) at the volume level.

The repositories are being created and populated automatically by a Python script, so if you notice any problems or strange happenings, please let us know either by opening an issue on the individual volume repository or by sending us an email. This is our first attempt at pushing
this data out. Please let us know what you think.

Available data includes:

Greek and Latin text generated by two open source OCR engines, OCRopus ( and Tesseract ( For work done optimizing OCRopus, see For work done optimizing Tesseract, see The output format for both engines in hOCR (, a format that contains links to the coordinates on the original page image from which the OCR was generated.

OCR results for as many scans of each volume of the Patrologia Graeca that we could find in the HathiTrust. We discovered that the same OCR engine applied to scans of different copies of the same book would generate different errors (even when the scans seemed identical to most human observers). This means that if OCR applied to copy X incorrectly analyzed a particular word, there was a good chance that the same word would be correctly analyzed when the OCR engine was applied to copy Y. A preliminary study of this phenomenon is available here: In most cases, the OCRopus/Lace OCR contains results for four different scanned copies while the Tesseract/AncientGreekOCR output contains results for up to 10 different copies. All of the Patrologia Graeca volumes are old enough that HathiTrust members in Europe and North America can download the PDFs for further analysis. Anyone should be able to see the individual pages used for OCR via the public HathiTrust interface.

Initial page-level metadata for the various authors and works in the PG, derived from the core index at columns 13-114 of Cavallera’s 1912 index to the PG (which Roger Pearse cites at A working TEI XML transcription, which has begun capturing the data within the print source, is available for inspection at: All figures are preliminary and subject to modification (that is one motivation for posting this call for help), but we do not expect that the figures will change much at this point. At present, we have identified 658 authors and 4,287 works. The PG contains extensive introductions, essays, indices etc. and we have tried to separate these out by scanning for keywords (e.g., praefatio, monitum, notitia, index). We estimate that there are 204,129 columns of source text and 21,369 columns of secondary sources, representing roughly 90% and 10% respectively. Since a column in Migne contains about 500 words and since the Greek texts (almost) always have accompanying Latin translations, the PG contains up to 50 million words of Greek text but many authors have extensive Latin notes and in some cases no Greek text, so there should be even more Latin. For more information, see

Next Steps

Developing high-recall searching by combining the results for each scanned page of the PG. This entails several steps. First, we need to align the OCR pages with each other — page 611 for one volume may correspond may correspond to page 605 in another, depending upon how the front matter is treated and upon pages that one scan may have missed. Second, we need to create an index of all forms in the OCR-generated text available for each page in each PG volume. Since one of the two OCR engines applied to multiple scans of the same page is likely to produce a correct transcription, a unified index for all the text for all the scans of a page will capture a very high percentage of the words on that page.

Running various forms of text mining and analysis over the PG. Many text mining and analysis techniques work by counting frequently repeated features. Such techniques can be relatively insensitive to error rates in the OCR (i.e., you get essentially the same results if your texts is 96% accurate or if your texts are 99.99% accurate). Many methods for topic modelling and stylistic analysis should produce immediately useful results.

Using the multiple scans to identify and correct errors and to create a single optimized transcription. In most case, bad OCR produces nonsense forms that are not legal Greek or Latin. When one OCR run has a valid Greek or Latin word and others do not, that valid word is usually correct. Where two different scans produce valid Greek or Latin words (e.g., the common confusion of eum and cum), we can use the hOCR feature that allows us to include multiple possibilities. We can do quite a bit encoding the confidence that we have in the accuracy of each transcribed word.

Providing a public error correction interface. One error correction interface already does exist and has been used to correct millions of words of OCR-generated Greek but two issues face us. First, we need to address the fact that we cannot ourselves serve page images from HathiTrust scans. HathiTrust members could use the system that we have by downloading the scans of the relevant volumes to their own servers but that does not provide a general solution. Second, our correction environment deals with OCR for one particular scanned copy. Ideally, the correction environment would allow readers to draw upon the various different scans from different copies and different OCR engines.

Posted in Uncategorized | Comments Off on Open Patrologia Graeca 1.0

Essays on Digital Classics and Digital Humanities

Gregory Crane
Professor of Classics and Winnick Family Chair of Technology and Entrepreneurship
Tufts University

Alexander von Humboldt Professor of Digital Humanities
Open Access Officer
University of Leipzig

Comments to

This link points a list and short summary of essays that I have written in 2014 and 2015 on Digital Classics and Digital Humanities. While most of these essays concentrate on Greco-Roman studies, I consider Digital Classics to include all Classical languages (and in practice I would include all historical languages, whether they are considered Classical or not) and the distinction between Greco-Roman and Classical Studies emerges as a periodic subtheme. The general theme of these essays is the challenge that professional students of Greco-Roman culture face in opening up their field so that it can serve the needs of our students and that it can participate in a global dialogue among civilizations.

Posted in Uncategorized | Comments Off on Essays on Digital Classics and Digital Humanities

“Privashing” vs. publishing — in search of an accurate term for a problematic academic tradition

Gregory Crane
Comments to

[Draft as of July 28, 2015]

Alexander von Humboldt Professor of Digital Humanities
Leipzig University

Professor of Classics
Winnick Family Chair of Technology and Entrepreneurship
Tufts University

This is a short piece reflecting on the need to distinguish publication that actually maximizes the degree to which we make academic work public from the traditional practice of handing control over to commercial, often for-profit, companies that make money by restricting publication. Of course, publishing under an open license has no necessary relation to any particular editorial workflow — researchers can, of course, carry out peer review or any other mechanism they choose. Certainly professors have wide latitude in deciding what does and does not count in a field and we can reward editorial work if we so choose. (I don’t know many academic administrators or members of the public who feel that we don’t produce enough publication and that we could not divert some of that labor to editorial work.) The point is that publishing should (in my view) mean that we have chosen that instrument that will help us advance the intellectual life of society as broadly and as deeply as possible.

I suggest, somewhat tongue-in-cheek, using the term “privashing”, instead of publishing for this traditional habit. Of course, it is an ugly term but that certainly suits the traditional practice in a digital age. The full (not very long) text is here.

Posted in Uncategorized | Comments Off on “Privashing” vs. publishing — in search of an accurate term for a problematic academic tradition

Revision to “The Big Humanities, National Identity and the Digital Humanities in Germany”

I have integrated a number of comments and clarifications from my colleagues in Dariah-DE and TextGrid in revising “The Big Humanities, National Identity and the Digital Humanities in Germany” — a good deal more has gone into making Dariah-DE international than had been clear to me when I first wrote this draft. I also have integrated information from Domenico Fiormonte’s essay, “Towards monocultural (digital) Humanities?” This essay outlines some of the problems of relying upon the Scopus database and rightly emphasizes the challenge of maintaining linguistic (as well as cultural and intellectual diversity). I have no good answers to this last problem except to say that we have our work cut out for us. For more on the Fiormonte essay, see here. This document remains subject to revision in light of comments and suggestions. — GRC July 20, 2015.

Posted in Uncategorized | Comments Off on Revision to “The Big Humanities, National Identity and the Digital Humanities in Germany”

The Work of Classical Studies in a Digital Age

Gregory Crane
[Draft as of July 20, 2015]

Alexander von Humboldt Professor of Digital Humanities
Leipzig University

Professor of Classics
Winnick Family Chair of Technology and Entrepreneurship
Tufts University

[This is hopefully the last essay, at least for a while, in a series that I have published on Digital Classics.]

Abstract: Who is the audience for the work that we professional researchers conduct on Greco-Roman culture? Frequently heard remarks, observed practices and published survey results indicate most of us still assume that only specialists and revenue-generating students really matter. If the public outside of academia does not have access to up-to-date data about the Greco-Roman world, whose problem is it? If we specialists do not believe that we have a primary responsibility to open up the field as is now possible in a digital age, then I am not sure why we should expect support from anyone other than specialists or the students who enroll in our classes. If we do believe that we have an obligation to open up the field, then that has fundamental implications for our daily activities, for our operational theory justifying the existence of our positions, and for the hermeneutics (following a term that is stil popular in Germany) that we construct about who can know what.

The full text is here.

Posted in Uncategorized | Comments Off on The Work of Classical Studies in a Digital Age

Resisting a monocultural (digital) Humanities

Gregory Crane
July 17, 2015

I would very much recommend reading Domenico Fiormonte’s piece “Towards a monocultural (digital) Humanities” (with reference as well to the plea by Miran Hladnik for linguistic diversity)” First, Fiormonte’s piece gives some very specific correctives to the Scopus data that I used when I wrote “the Big Humanities, National Identity and the Digital Humanities in Germany” a month ago. Second this piece articulates, in its title and its content, the fraught question of how we support linguistic and cultural diversity in a transnational space. The question before us is, I think, how we maintain our various linguistic and cultural identities while communicating with one another in a dynamic and enriching dialogue across boundaries of language and culture.

For more on this, look here.

Posted in Uncategorized | Comments Off on Resisting a monocultural (digital) Humanities

Greek, Latin, and Digital Philology in Germany and the United States (part 2):

Part 2 of 2 (Part 1 is available here.)
Greco-Roman Studies in the US
July 14, 2015

Gregory Crane
Alexander von Humboldt Professor of Digital Humanities
Leipzig University

Professor of Classics
Winnick Family Chair of Technology and Entrepreneurship

Summary: [The full text is available for comment here.]
I have now released a draft for part 2 of Greek, Latin, and Digital Philology in the United States. This part includes some information about Greco-Romans studies in the US, with some comparisons with the situation in Germany, and then moves on with a very brief and preliminary start for suggestions as how Germany can make itself an (even more) attractive location for a research career in this field.

Tables 20 and 21 address the basic size of Greco-Roman studies in the United States. There were, according to one survey, 276 departments of Classical Civilization in the US in 2012, with 1,410 tenured or tenure track faculty. There are 276 US departments of Classical Civilization, while the 52 universities that have chairs in Greek, Latin, Ancient History or Classical Archaeology would be equivalent to 208 departments (if Germany had the same proportion of universities and had a population of 320, rather than 80, million). But even if we factor in the differing populations, the 200 chairs for Greco-Roman studies in Germany are only equivalent to 800 in a US-sized population, whereas there are 1,410 tenured and tenure-track positions in Departments of Classical Civilization in the US. In absolute terms, the 290 tenure-track positions (presumably assistant professors) outnumber the 200 chairs in Germany. A Professor Doctor in Germany is different, of course, than an Assistant Professor who still needs to earn tenure but the American system offers more points of entry into the tenure system than there are chairs in Germany. There are, I think, a good number of middle level positions in Germany but most of these positions offer a guarantee: after six years, you’re out and you need a new job. Bad as the the long term job market is in the US, it looks a lot better to me when I look closely at the situation in Germany.

Tables 22-25 attempt to identify the business model upon which Greco-Roman studies depends in the United States. Table 22 clearly identifies at least one feature upon which Greco-Roman studies does not materially depend: there are only 1.6 graduating seniors per faculty member (perhaps 5 majors, assuming a few second semester first year students declare per faculty member). Anyone who teaches in a US Department of Classical Civilization knows that larger classes, aimed at non-majors, provide the basis upon which we depend to justify our positions. I have, however, found no statistics on the size of these courses overall — and this deserves a major study if we we want to understand the current health and future prospects of Greco-Roman studies in the US.

At the same time, the Modern Language Association (MLA) (Tables 23-25) provides us with statistics for enrollments in Greek and Latin: there were in fall 2013 still 40,109 students reportedly enrolled in courses of Greek or Latin — 28.4 such students for each of the 1,410 tenured and tenure track positions. We need to be cautious in assessing these numbers — there are almost twice as many institutions that reported enrollments in Greek or Latin as there are departments of Classical Civilization (the MLA states that 512 institutions reported enrollments in Greek and/or Latin but the 2012-13 Survey of Humanities Departments at Four-Year Institutions by the American Academy of Arts and Sciences (AASHD) identified only 276 departments of Classical Studies), but even if we assume that half the students of Greek and Latin are in institutions without departments of Classical Civilization, we get about 15 students of Greek and Latin for every tenured and tenure-track professor. This reflects a discipline-wide commitment to keeping the study of the languages alive.

The MLA numbers also told two stories. First, there was a precipitous drop in enrollments between 2008 and 2013 — about 20% for both Premodern Greek and Latin (when different ways of classifying Premodern Greek are taken into consideration). I think that this surely reflects anxiety about the practicality of undergraduate study after the financial crisis of 2008. Whether we can reverse these losses or whether this is the new normal remains to be seen. But if we consider the figures from 1968 through 2009, we see substantial (to me, amazing) resilience: despite the crises and changes that followed the 1960s, there are about as many people studying Greek and Latin in 2009 as there were in 1968. This was a huge achievement and something for which the study of Greco-Roman culture in the US should take pride. I do think that we will need new ideas and new methods to maintain this resilience but I personally think that we are poised to grow and expand if we are determined, fearless, and judicious. We are poised to reinvent the study of Greek and Latin at every level — but that must remain, for now, an assertion and await another venue for further discussion. More than 75% of all historical language students in the US study Greek or Latin (Table 26) — if smaller historical languages (e.g., Aramaic, Akkadian, Sanskrit, Classical Chinese) are to flourish, the students of Greek and Latin must design a general infrastructure that serves many other languages as well.

Table 27 turns to question of where tenured and tenure-track professors of Classical Civilization in the US got their PhDs. I analyzed the public web pages for 575 US Assistant, Associate, and full Professors in this field. Among 206 faculty at institutions without a PhD program, the national composition was very similar to the Professor Doctors of Greek, Latin, Ancient History, and Classical Archaeology in Germany. In non-PhD departments in the US, 95.6% of the faculty (198 of 206) had US PhDs, while 95% (190 of 200) of the German chairs had PhDs from German institutions. When we considered PhDs from other Anglophone and German-speaking universities, we accounted for 98% of the faculty in both the US (203 of 206) and Germany (196 or 200). If you want to become a Professor Doctor in Germany or a tenured/tenure-track Professor at a non-PhD US program, you had better get a PhD in the US or Germany. You might get one of these positions if you get a PhD in an English-language or German-language program but I would not count on it.

If we look at the departments of Classical Civilization with (by one ranking: the top-10 PhD programs, we find a very different population. Just under two-thirds of the Assistant, Associate and full Professors in these departments received their PhDs from US programs (64.5%, 102 out of 158 faculty where I could determine the PhD institution) — adding the three Canadian PhDs would get us to almost exactly two thirds (66.5%, 105 out of 158). Thus, fully one third of all these faculty received their highest degree (there was one faculty member who seems only to have received an MA) outside of North America. Most of these (33 out of the overall 158, 21% of the total) came from the UK while two came from Australia.

More than 11% (18 of 158) of these faculty received their PhDs from outside the Anglophone world. With 10 departments, this means that each department has, on average, one or two faculty members who were trained outside the Anglophone world, reflecting a very different scholarly tradition and (often) maintaining deep ties with colleagues in the nations where they were trained. For me, the importance of such international faculty cannot be overstated — when I was a student, I benefited constantly from working with faculty who had not come through the US system. Some may view the fact that fully one third of the faculty at the highest ranked departments do not have US PhDs as a sign of weakness — there are not, in this view, enough good Americans to fill the positions. I see this diversity as a strength of the US system. This strength may only be practical because the highest ranked departments are also the biggest and each can afford to take a chance on one or two faculty who might not necessarily flourish in the US system (I know of at least one instance where a big department brought a big scholar in, knowing he would never fit in — they felt they could afford it).

Table 28 looks quickly at gender balance. The American Academy of Sciences report (from which many of the data are drawn) reports that 40% of the Classical Civilization faculty are women while women accounted for 38% of 582 US faculty members whom I analyzed. The rate for full Professors is lower — 33% — but that 33% is still 50% higher than the 22% of female Professors Doctors in Germany.

The final table (Table 29) summarizes where the faculty I identified got their PhDs. I was most interested in the rates for Assistant Professors — PhD programs have changed substantially since current Assistant Professors chose where to get a PhD, but those departments have changed even more since most Associate and full Professors got their degrees.

The final section provides some partial, preliminary, and perhaps provocative comparisons between Germany and the US in Classical studies. Any student, with a choice of beginning their career in Germany or the US and who can manage either German or English, should consider the following: data reinforces the more general impression that English language scholarship no longer cites non-English scholarship at the same level as even a generation ago; there are more permanent jobs in the US; the most highly ranked departments have between 15 and 22 faculty members and are, arguably, better suited structurally to support a more generalized Altertumswissenschaft; if the student does manage to get a tenure track job (no easy task), then that person immediately becomes a critical member of a(ny rational) department; there is very little evidence that people from outside the German speaking world are going to win one of two-hundred or so coveted chairs in Greek, Latin, Ancient History and Classical Archaeology in Germany.

The full text is available for comment here.

Posted in Uncategorized | Comments Off on Greek, Latin, and Digital Philology in Germany and the United States (part 2):

Greek, Latin, and Digital Philology in Germany and the United States

Part 1 (of 3)
Gregory Crane

Alexander von Humboldt Professor of Digital Humanities
Leipzig University

Professor of Classics
Winnick Family Chair of Technology and Entrepreneurship

I have published some data in a summarizing essay about Greco-Roman studies in Germany. There is a semi-autobiographical opening that can be skipped and a series of 19 tables that can be quickly skimmed. The essay has a very particular point of view: as an Alexander von Humboldt Professor in Germany, my job is to help make Germany as attractive a center for research as possible. I thus begin by exploring the use that my American colleagues make of German Greco-Roman scholarship. The paper published so far is primarily descriptive, invites comment and seeks to lay partial groundwork for further analysis and possible suggestions.

Table 1 basically says that if a Humanist wishes to publish in German and attract an American audience, Classis seems to be the field to do it. Students of Greco-Roman culture cite (and presumably read) German.

Tables 2-4, however, present preliminary but suggestive evidence about the widely perceived decline of German scholarship in English-language publications on Greco-Roman Studies. In the 1956 and 1957 issues of the Transactions of the American Philological Association, only 48% of the citations pointed to English sources and 32.5% citations pointed to German. By 1986, the percentage of citations to German had fallen to 24% (it was 22.4% in the 1986 American Journal of Philology). By the 21st century, German citations had dropped to c. 10.9 in AJP 2002, 8.5% in TAPA 2006, 8% in AJP 2014, and 10.9% in TAPA 2014. For a quick (and obviously preliminary) comparison with a British journal, I analyzed the first ten articles of Classical Quarterly 2015 and found that 10.3% of the citations pointed to German. Where I had done the analysis myself and had the data, I was able to calculate the percentage of German scholarship among cited publications that were thirty years old or less. In AJP 1986, 17% of citations thirty years old or less pointed to German. In AJP 2014, TAPA 2015 and the first ten articles of CQ 2015, I found that 4.7%, 8%, and 6.1% of citations pointed to German. The numbers behind these statistics are not large but the fact that the results are so consistent is striking.

Tables 5-7 look at the extensive, even brutal, cuts to which German chairs in Ancient Languages, Literatures and Cultures have been subject. Between 1997 and 2011, the number of chairs for Greek declined from 39.5 to 32.5 — a loss of 7. Latin declined from 57 to 48.5, a loss of 8.5. And Greek and Latin were comparatively lucky — Indo-European lost 8.5 of 22 chairs, Egyptology 3.5 of 18 chairs, Assyriology 5 of 21. Ancient History fared relatively well (losing only 2 chairs and declining from 79 to 77), but Greco-Roman History plays a crucial role in training primary and secondary school teachers and a German report on Small Fields suggests that this is responsible for Greco-Roman History’s survival in this form.

Table 8-10 provide some additional background. Of these table 9 is worth noting: even as Greek and Latin lost chairs from 1997-2011, the number of students was increasing. The German universities managed this by increasing the number of non-Professorial positions. The number of students per professor rose from 36.5 in 1997 to 46.1 in 2007 to 56.2 in 2011, while the number of students per academic staff remained steady at about 10 (9.9 in 1997, 10.8 in 2007, and 11.1 in 2011). This may indicate an institutional shift away from a research orientation and towards a service orientation for Greek and Latin philology and a corresponding judgment on the perceived value of research in this field.

Tables 11-13 describe how often current professors received their Phds from German universities, German-speaking universities, and from elsewhere. For Greek and Latin Philology, Ancient History, and Greco-Roman Archaeology, 190 of the 200 (95%) professors that I could identify had German Phds, 6 had PhDs from Austria or German speaking Swiss universities. Thus, 98% of German chairs had received their PhDs from the German speaking world (or the Deutschsprachiger Raum, as we say in these parts). Of the remaining four, two were Germans who had received Oxbridge degrees and two were from countries adjacent to Germany (Denmark and the Netherlands). No one with a PhD from outside of Europe holds any of these chairs. I must admit that while these numbers did not surprise me, actually seeing these numbers so starkly defined left me personally a bit shaken. Table 13 presents some figures for international students: 18.6% in Greek (110 of 592), 15% in Classical Philology (89 of 591), and 10% in Ancient History (36 of 345) — Latin has only 3% (125 of 4,268) but that probably reflects the fact that most students of Latin aim to teach in the German school system. Of the 200 professors, I could only find two (one from Spain and one from Greece) who had moved from their home country to study in Germany. Thus more than 10% of the students are international but perhaps 1% of the professors started as international students.

Table 13 lists where these German professors most commonly received their PhDs: Heidelberg, Freiburg, Munich (LMU) and Cologne emerge as the big four, accounting for 40% of all chairs (80 or 200).

Tables 14-17 look at gender balance. The figures are low — 22% of professors are women — but this number also reflect the percentage of all academic staff in Greco-Roman Studies in 1997 (22.8%). In 2013, the percentage of women among the academic staff had, however, risen to 39.7% and the percentage of female professors may well increase accordingly in the future.

Finally, tables 18-19 report an analysis of the relative percentage of publication in German vs. English (based on looking at up to 10 publications on the public websites of each professor). Greco-Roman Philologists and Archaeologists publish 90% in German and 10% in English, while Ancient Historians publish 97% in German and 3% in English. (I undertook this analysis to see if the decline in German citations reflected a shift to publication in English.) By contrast, in Egyptology, 71% of the publications are in German and 29% in English while in Assyriology 73% are German and 29% English.

Full text at:

The full text is here.

Posted in Uncategorized | Comments Off on Greek, Latin, and Digital Philology in Germany and the United States

Where did holders of German Chairs in Greek, Latin, Ancient History and Classical Archaeology get their PhDs?

Gregory Crane
June 28, 2015

As a by-product to a larger study of Greco-Roman studies in Germany and the United States, I have published some figures on where faculty from ranked PhD-granting departments got their PhDs.

For the data, click here.

Posted in Uncategorized | Comments Off on Where did holders of German Chairs in Greek, Latin, Ancient History and Classical Archaeology get their PhDs?

Update: Where did faculty in US Classics Departments with top-ranked graduate programs get their own PhDs?

Gregory Crane
Leipzig and Tufts Universities

Maryam Foradi
Leipzig University

June 25, 2015

[For the earlier version, see:]

On June 11, we published a discussion of the programs from which the top ten departments in a particular ranking of Classics PhD programs chose their faculty. We here publish an analysis based on all thirty-one departments in that ranking. We have sorted the results based upon numbers of assistant professors — this reflects activity over the past decade or so (if we consider both the PhD training itself and the period that assistant professors have served).

The statistics do change a bit. Four PhD programs — all in the US — now stand out, each having placed 6 or 7 of 65 assistant professors in these 31 programs and, altogether, accounting for 40% (26 out of 65) of all the assistant professors in this group. A second cluster of 11 programs (including Oxford and Cambridge) placed 2 or 3 of their PhDs in these departments. A third cluster of 12 programs (8 of them situated outside the US) placed one of their PhDs in these departments.

For the full update, check here.

Posted in Uncategorized | Comments Off on Update: Where did faculty in US Classics Departments with top-ranked graduate programs get their own PhDs?