Unfortunately, linguistic injustice matters.

In a previous letter [1] I argued that the absolute dominance of the English language in scientific communication has led to linguistic injustice, arising for two main reasons: (i) non-native English-speaking scientists (NoNES) support all the costs of having a English as a common scientific language; and (ii) while native English-speaking scientists (NES) benefit for free from having a common language, they are in an advantaged position in any scientific discussion due to linguistic skills. Guariguata et al. [2] replied that there is no sharp difference betweenNoNES andNES, basically because the more one studies and practices English, the better his or her English will become. Of course, I agree with this, although I think there is no relationship between this statement and the linguistic injustice problem. The fact that some NoNES, with a lot of work and great merit, are able to overcome linguistic barriers does not imply that the injustice situation does not exist or is unimportant. Guariguata et al. [1] do not see the associated costs to NoNES derived from linguistic injustice as a big problem nor do they believe that they can result in any important bias in the research output of scientists (although, surprisingly, they do propose solutions to overcome it). However, the available evidence highlights the role of linguistic difficulties of NoNES as a source of publication bias. Primack et al. [3] analyzed the results of the review process for manuscripts submitted to Biological Conservation and found that ‘‘there is substantial disadvantage of not having English as a first language; this disadvantage could result in a decrease of around 30% in the chance of acceptance’’. Vasconcelos et al. [4] showed that the research output of Brazilian scientists was tightly related to their Englishlanguage skills and Man et al. [5] found that English proficiency was a good predictor of national research production, even better than the investment of each country in science. I found that the proportion of papers published exclusively byNoNES in ecological journals was negatively