Escrevendo cartas de recomendação ou referência

Antes de tudo, leia Advice to Graduate School Recommendation Letter Writers, de Shriram Krishnamurthi.

Você só deve escrever cartas para pessoas que tenham trabalhado com você, ou cujo trabalho você conhece bem. Escrever cartas para pessoas em outra situação é muito ruim porque a carta vai ser fraca, e você terá que escolher as palavras com muito cuidado para não prejudicar a pessoa além do que ela já vai ser prejudicada por ter escolhido uma pessoa distante dela como referência.

Para cartas de promoção, que são tipicamente solicitadas por universidades americanas, resumo a seguir dicas de três colegas de lá que me ajudaram muito na época que tive que escrever uma carta dessa:

• Normalmente não tem problema se você é co-autor de trabalho com a pessoa, ou se tem outra conexão. Só informe isso logo no início da carta (“I was co-author with Y on papers A,B,C”). Mas é importante confirmar que isso não é um problema (conflito de interesse) logo que a universidade lhe procurar para escrever a carta.
• Duas páginas é um bom tamanho. Um pouco mais se é alguém que você quer realmente promover.
• Leia e resuma o principal diferencial dos melhores (3-5) artigos da pessoa em questão.
• “My general outline would be something like: I adjective support giving person tenure at your university. For background I am Paulo Borba from explain. My relation to person is that explain. Discuss the papers they sent or at least the person’s work from their CV or your own knowledge. In sum because of the reasons explained above I adjective support giving person tenure. you can also say what you would do at your own university in this case or compare to others Please contact me if you have any questions.”
• “Best practice is to stick to an evaluation of the quality of the candidate’s contributions to the field, instead of looking at their numbers (publications, funding, etc.).”
• “I most like it when the reviewer can demonstrate some understanding of the candidate’s work. When I write letters, I pick out 2-3 papers to (at least lightly) (re-)read, and explain what I like about those papers. What problem did it solve? Why was it worth doing? What was clever/interesting about how they did it? What impact has it had? That sort of thing. Anywhere from 1-3 paras per paper, nothing longer than that.”
• “At any rate, what I expect to see when someone is going to get tenure is that they have established themselves as independent leaders of a field worth pursuing. That means:
  • established (not just started)
  • independent (they can have collaborators who are of equal or greater seniority, but then must have also done some work without those collaborators, only with younger researchers)
  • leader (doesn’t necessarily mean “#1”, but acknowledged as someone whose opinion on the topic people care about)
  • field (this can of course be “manipulated”: make it small enough and they may be leader by virtue of being the only one, make it big and they don’t have a chance)
  • worth pursuing (hopefully self evident)”
• “A different way to think about this is, this person will have lifetime employment. The expectation is that they will be using that freedom to have a successful career with meaningful, impactful work. What evidence do you have that they will indeed achieve that? The above is one very clear, concrete way of demonstrating why you have such a hope, but maybe there are other interesting (maybe less standard) ways also.”