I've been living with a split personality disorder for many years. Depending on who's asking and why, I'm an "ETH assistant", a "PhD student" or, sometimes, simply a "researcher". Until now nobody has been able to give me a consistent definition of what a "Doktorand" – as this position is called at ETH – really is. So I went into the matter myself.

As far as the Rector's Office is concerned, I'm a student – holder of a student identity card and a "Testatbogen" (attestation of presence at lectures), which is as it should be. I find the former very useful, not only because it entitles me to special low prices in the canteens and mensa but also reduced ticket prices to a wide range of cultural activities in Zurich. But somehow the "student" label doesn't describe my existence very accurately. I haven't been to a class for a long time; I'd feel like a wrinkly if I wandered into the annual freshman party, which is why I leave the hangovers to the undergraduates. And suddenly, marriage is a topic amongst my friends. I don't feel old exactly, but I definitely no longer fit the typical profile of a student.

At the end of each month, always a good-news day, I get mail from the ETH personnel department with a salary statement advising me of the transfer of a stately wage. In contrast to Daddy's allowance check, this wage enables me to live well in Zurich, while the deducted social contributions give me the feeling of belonging to the "grown-up" working portion of society. As far as the personnel department is concerned, I'm an assistant, salary grade 15 with a 100% position – they don't even know that I'm studying for a PhD. But my status still can't be compared to a position in the private sector. In the usual run of things at ETH, unpaid overtime is only accepted, at the best, for those further up the ladder. Once a year the personnel department sends me the obligatory letter "terminating" my contract, and, above all, for the next three years I'm entirely dependent on my boss as to whether I grow or perish as far as my job and future personal development – viz. my doctoral thesis – is concerned.

I don't want to fall into a whingeing orgy here. If a graduate student is on good terms with the professor supervising his or her thesis and lucky enough to work at the "right" institute, then ETH provides almost ideal conditions in many respects; first class academic development coupled with financial security.

Martin Näf, graduate student of computer science and current ETH Life columnist.

However, where trust between the two parties involved is missing, the system can very quickly turn inside out. Suddenly, a secure existence is out of the question and, in contrast to the undergraduate system, there is a total lack of structure and, above all, clear and explicit criteria for success. The fact that graduate students in these cases are often treated like freshmen, but with the expectation that they will please continue to spend weekends in the lab instead of going to parties, exacerbates the situation.

There are efforts to give more structure to the status of graduates. Tying graduate studies to the credit system is already reality, but transferring this into practice is often combined with calls for stronger "formal schooling". Just how this fits in with the goal of a mature scientist – who is not merely a specialist – at the end of graduate studies is something on which opinions are divided. My own experiences with graduate students I met in the United States – not just the few high achievers from Stanford or MIT – strengthen my conviction that more formal schooling for graduates does not necessarily lead to the wished-for creation of independent scientists. A mere copy of the Anglo-Saxon system would certainly not nurture quality, but above all it wouldn't fit into our system of inherently unstable positions and a practically non-existent scholarship system. But there are no easy solutions for the conditions of employment. A guaranteed position for a number of years would prevent the effective removal of the occasional "bad egg". The fact that unlimited work contracts at ETH can only be terminated following absolute incompetence or massive transgression is incompatible with the level of work that is called for.

So what other possibilities are there? There is hardly a substitute for the principle of "trust", but this is something that has to be worked at continually and continuously by both parties to a contract. Supervisors who take their responsibility seriously and don't hold ignorant views, like "it's-always-been-done-like-this" or "others-are-worse-off" are indispensable. Ensuring the presence of such paragons begins with the appointment of professors, and measures need to be put into place to reward or sanction the subsequent fulfilment or non-fulfilment of this duty. As long as the list of publications is the main criteria for appointment, long-term, responsible behaviour cannot be fostered. Results of the ongoing personnel surveys could provide a possible basis for evaluation, which it is to be hoped, will be used. But we graduate students must also do our part by actually claiming the support we require, and completing it with our own initiative. This is the only way to make my preferred definition of a graduate student reality, namely an independent researcher with initiative and ambition.