Graduate School in Japan

On being a PhD student at the University of Tokyo

I am frequently asked about my experience studying in Japan and I finally decided to write and upload a short essay. I have also received many emails about how to get admitted to the University of Tokyo as a PhD student. While I do not mind at all getting in touch with the writers of those emails, I figured that answering some of the most frequent questions in advance might help everybody.

Tokyo University's Hongo Campus, with Shinjuku's skyline and beautiful Mt. Fuji in the background (photo taken by Martin Wenk)
Tokyo University's Hongo Campus, with Shinjuku's skyline and beautiful Mt. Fuji in the background (photo taken by Martin Wenk)

Let me begin with an unsurprising confession. Obviously, before even starting my degree, I also got quite excited about going to Tokyo and travelling around in Japan during my studies at the University of Tokyo. While studying and working hard was very important to me, I was also extremely keen to explore the country and to learn about the culture and the language. Studying at the University of Tokyo was a highly rewarding experience for various reasons (among which the possibility to eat fantastic sushi every day cannot be neglected ;-)) but not necessarily the way you would expect it. Studying Computer Science in Tokyo may not be the standard way to obtain a PhD degree (I doubt that the standard way even exists but if it does it is probably not in Japan - the Japanese have their own «standard» way). It is arguably not the world's hardest way either, since both admission and graduation are probably not as competitive as their counterpart at a top school in North America. However, since the research world as a whole is rather competitive, independent of where you are based at, having a lower level of local competition can make an early research career even more difficult at a place like Tokyo, where many of your fellow Japanese are not as open about their competitiveness as an «average» Westerner would be. At the beginning, this lack of passion (or at least this is how it seemed to me) of many of my fellow students made me feel a bit depressed; I was also tempted by the possibility of just getting an «easy» degree until I realized that such a path would not be satisfactory for me at all. I became to believe that the outcome of my studies really depends on my personal initiative and on what I make out of my time in grad school. I ended up learning many things about myself (which I hope every grad student does) but I also learned a lot about myself in a completely different environment. I learned about different styles of and approaches to research and collaboration - to an extent that probably would not have been possible at an American or European school. While the University of Tokyo may be considered an international graduate school, most international students are from Asia and their way of thinking was often different from mine.

Grad School Admission - How to become a PhD student at the top university in Asia

Graduate school admission at the University of Tokyo may be a bit different from admission at universities in North America or Europe. It is not necessary to obtain a Master's degree at the same school. I actually got my Master's in my home country, Switzerland. It is however key that you do have an advisor at the time you apply. Although it is the graduate school deciding whether you will be admitted or not (at least formally), your future advisor will have a very important say in the decision.

The application is paper-based, consists of many pages, and a careful completion is absolutely crucial. I would advise you to get help from a native Japanese speaker to fill out all the parts. While writing in English is allowed for most of the contents, handling Japanese forms can be rather challenging. Arguably one of the most important parts in the application for the PhD course is a research plan, describing what you want to investigate during the next three years. On the web, there is plenty of advice on how to write a research plan so I will not elaborate on this part. Other than the research plan, there is no graduate school essay or similar, but mostly just lists of the schools attended, your previous publications, etc. Completing the TOEFL test (TOEIC may be acceptable) is required (I have heard that it is mandatory even for native speakers - I hope that this requirement was dropped in the meanwhile). Once the application material is accepted, the next step is to take the entrance exam.

Entrance Exam

The following information applies to the entrance exam of the Computer Science Department (which is part of the Graduate School of Information Science and Technology) only.

The entrance exam consists of three parts:

In case you already prepared for standardized tests such as the GRE and the GRE subject test or in case you attended a very good college, you should be able to pass the entrance exam without too much of an additional effort.

Graduate School Scholarships - how to finance your studies in Japan

Before going to Japan, I thought about working at a sushi restaurant to finance my studies but once I realized that I would probably never even get to touch the rice (not to mention the fish) during the three years, I dropped this naive plan. Being admitted to a graduate school at the University of Tokyo does not necessarily come with funding. Most likely, you will have to either pay school fees (not that expensive compared with a top private school in the US) and living expenses (it's Tokyo after all!) out of your pocket or apply for a grad school fellowship. Such applications usually consist of a scholarship essay in addition to documents similar to the ones you completed when applying to the department. There are many scholarships and fellowships and not that many applicants (from the West) but obtaining them is still competitive. I was very fortunate to receive a University of Tokyo Fellowship for the second and third year of my PhD, in addition to financial support from my professor.

Life as a PhD student at the University of Tokyo

Let me quote Robert C. Christopher (excerpt from The Japanese Mind, p. 85 "The Education Race")

At the very top of the heap is Todai - the acronym for Tokyo University - which in Japan carries the prestige of Harvard, Yale and M.I.T. all rolled into one. When a Japanese admits to you that he is a graduate of Todai, he generally does so, in my experience, in a flat, noncommittal tone, as if fearful that you will think he is putting on airs. And the admission is, in fact, inevitably a kind of boast. [...] In short, if you are a Japanese parent and want your child to be a winner in life, you do everything possible to ensure that he or she gets into Todai [...]


In the Computer Science Department, the requirements in terms of credit points obtained by actually taking classes are rather low. While every student must attend the seminars of their own working group (can be quite challenging since most of the presentations tend to be in Japanese, more on language challenges below), there is only a low course load on top of these seminars. You are required to take one (sic!) class during the three years of PhD studies (recall that you must have a Master's degree already, course work is considered «done» at the time you enter the PhD course). Since most classes are in Japanese as well (due to the Global 30 program «Thirteen Universities to Lead Japan's Internationalization», an initiative to foster internationalization in Japanese universities, there may be English lectures soon, I actually co-taught one English lecture myself) I decided not to take any class in my first year. I also anticipated that I would be busy writing my thesis in the third year. The second year would thus be ideal to take the class required. Although I studied Japanese very hard during my first year, I must admit that my ability to understand the precise meaning is still somewhat limited. While I can easily carry on a casual conversation, I still have a hard time understanding all the words necessary to follow a technical or scientific discussion. In the class I took, I may have been rather lucky with the professor: he offered to write on the blackboard in English, while talking in Japanese. He was also speaking rather slowly, repeating the most important parts several times, such that even a foreigner with limited language skills could understand the most important points of the subject. I do not know whether he was just trying to accommodate the foreign student or whether Japanese classes are generally held at a rather slow pace. Class room activities were very limited: no questions were asked, the interaction, if there was any, was hardly observable. In order to obtain credit points, each student needed to submit a report at the end of the semester. The assignment was a not-so-hard programming exercise that involved a couple of hours of coding and experimenting. Due to conferences etc. I was out of town for a couple of weeks during the semester. While I got away with it without any problems for the class I took, for some other classes, attendance is actually mandatory (even at the graduate level).

Research («earn» your degree)

In my opinion, research should be the most important part of studying towards a PhD degree. Although being highly relevant, treating this topic would go well beyond the scope of this essay. There is a lot of good and helpful advice on how to do research available online. You will have to find your own way of doing research using these sources and, more importantly, your advisor(s) (in Japanese, your sensei) and more senior students and researchers (in Japanese, they would be called sempai - relationships between sempai and kouhai would arguably deserve their own article as well).


During my graduate school studies at the University of Tokyo, it has been difficult at times to find collaborators satisfying all of the following requirements (some of these may be general, others are rather specific to Japan): interested in what I wanted to investigate, willing to collaborate (in particular with me / a foreigner), and sufficiently fluent in English. Keep in mind that it takes quite a lot of time and effort to build research relationships and that three years is a very short time to establish a fruitful collaboration.

Language and Culture

My personal perception was that many Japanese students do not feel comfortable to speak (up) in English. Apparently, students would communicate in English and German a hundred years back, let me quote T. Philip Terry (Terry's Guide to the Japanese Empire, p. 191, 1928 edition, written around 1900)

The Imperial University (Teikoku Daigaku) of Tokyo occupies a group of semi-classical buildings a short way S. W. of Uyeno Park, in Hongo-ku in the extensive grounds of the one-time metropolitan residence (of which the old gate is the only remaining relic) of the daimyo of Kaga Province. Tram-cars trun to within a short distance of the main gate, which is always open to those interested. Almost every one about the place speaks English, and not a few speak German, since Teutonic minds have left a strong impress upon the institution.

Almost every one about the place speaks English - somehow, this appears to have changed in recent years.


Research-wise, being at a place like Tokyo can be very challenging and demanding, requiring a high degree of independence and perseverance. Despite (or maybe because of) all these challenges, I would certainly recommend a prolonged stay in Japan. Doing my PhD at the University of Tokyo has been a great experience. It has also provided me with the unique and wonderful opportunity to, while getting my PhD, also learn a new language and culture, and to expand my horizon.



Although grad school rankings arguably should not be the most important factor when making a decision on which school to go to, they may certainly have an influence. In case your scientific career does not take off, the following information (which, whether justified or not, many recruiters do consider) may be relevant as well: graduate school rankings and university rankings such as World's Best Universities: Top 400 - US News and World Report, the Academic Ranking of World Universities, and Times Higher Education - QS World University Rankings regularly rank the University of Tokyo among the top schools worldwide and as the number 1 in Japan and in Asia.
SourceYearWorldRegion (Asia)Country
US News20102211
Times Higher Education20092211
Times Higher Education20081911

Universities that are often ranked higher than the University of Tokyo are institutes such as Harvard, Stanford, MIT, Princeton, Yale, Cornell, CalTech, Cambridge, Oxford, University College, Imperial College, Chicago, Columbia, Penn, Johns Hopkins, and Duke.

What the University of Tokyo writes about itself

Extracted from the website of the Global 30 Program. The following is a quote.

The University of Tokyo - A leading center of knowledge in the world

Having celebrated its 130th anniversary in 2007, the University of Tokyo, the first and oldest national university in Japan, has been a leading center of knowledge in the world. The University of Tokyo is conducting a wide-range of research in humanities, social sciences, and natural sciences, while making a constant effort to create an interdisciplinary and cutting-edge academic curriculum by building upon inherited traditions and expanding fundamental academic disciplines. The University of Tokyo is renowned for its Nobel Prize-winning research. Nobel Prizes have also been awarded to the University of Tokyo graduates for Physics, Literature and Peace.

Information in German

In October 2009, a European career website published a series on studying in Japan. An interview with me (executed by email) was part of the coverage.

About the Author

Christian Sommer investigates algorithms and graphs in English, loves and eats in Japanese, drinks wine in French, and plays in his native (Swiss-)German.