Tony Marmo http://tonymarmo.tripod.com/linguistix-logik/ wrote: Let me shift the issue a little bit. In Generative Linguistics done in Brazil there is one ideological bias that is undeniable: the segregationist bias. This idea is not something Brazilians came with. It started with foreigners that began to study 'Brazilian' Portuguese. They were people whose native language was not Portuguese and whose academic background was rooted in the 'classic letters'. People who just jumped from classic Philology to Generative Grammar and who were strongly influenced by the work done by the Summer Institute of Linguistics in the 'field' of Indian languages until the 80's. Caesar's motto, 'divide and conquer', was applied and the indoctrination spread. Firstly, whenever you found any new village of ten individuals, they could not speak any dialect of any known Tupian language. They had to speak their own language. Within time, this had to be applied to the major 'non-Brazilian' language, Portuguese. We could not speak the same language as the Portuguese, anymore, so told us these foreigners. And every detectable difference between the dialects spoken in the one or the other country had to be over-exaggerated. The next step was to deny the similarities. Even if you could find identical sentences, the indoctrination continued saying that the sentences were only identical in the 'surface structure', but they had to have different 'deep structures', regardless of what theory you used. Of course, there is no such thing as a uniform language in Brazil or Portugal, and it is not only the case that there are only two varieties of Portuguese. But the idea of dialectal continuum had to be dismissed to keep the dogma that there were two languages and not several dialects of one language, so that Brazilian internal regional differences had to be considered unimportant, although evident, and internal regional differences in Portugal had to be declared 'non-existent'. For years, Brazilians have listened to completely untrue 'features' ascribed to their dialects, such as the infamous idea that we did not speak a pro-drop language anymore, and have repeatedly provided examples to the contrary that have been systematically ignored by the segregationists. Nowadays, the 'segregationist' indoctrination has dominated the academic scene, and whenever someone says the magic word, 'Portuguese', there is always an indoctrinated person to ask 'which Portuguese?' The problem is that the 'segregationist' bias is so strong that the segregationists do not think they have a bias or a prejudiced view. On the contrary, they think that the others' views are biased. Their bias only becomes evident by their almost irrational reactions when someone dares to challenge their preconceptions. The attitude is perfectly explainable: the segregationist linguists do not want others to think differently. They want to rule out the possibility of anyone talking about Portuguese as one single language, of anyone dissenting from them in one key issue. And they have virtually succeeded in doing so. Thus, coming back to your point, the question: (Q) Do the mass media have a liberal bias, or is it the case that those, who accuse the mass media of being liberal, have a conservative bias? cannot be answered through a statistical. The point is to determine which faction wants the cultural elimination of the other, and which is prepared to accept the coexistence. Thus, one has to ask what those, who claim that the media have a certain bias, want to achieve? Scott Martens http://pedantry.fistfulofeuros.net responded: Tony, while I largely agree with you about the generativists quite incoherent ideas about the social functions of language and their lack of critical reflection on their understanding of language/dialect/idiolect distinctions, and while I have similar criticisms of SIL (although their mandate is rather narrow and their village-by-village conception of language has some justification for them), I have to confess that I think the Brazil/Portugal linguistic distinction is at least as big as the US/UK distinction and probably bigger. I can understand spoken Brazilian Portuguese without too much trouble, but after moving to Europe it took me a month to figure out that I got RTP International on cable. Until I had watched the channel long enough to start seeing on-screen text, I was sure that they were speaking Polish. I've never had this problem in Spanish - even Bolivian Spanish, which is just weird - and I've met a lot of Europeans who've had this sort of trouble with Canadian French. But the only time I ever had comparable trouble in English was in Scotland, where I once had to ask someone if they spoke French because I couldn't understand a word they were saying. So, the shoe goes on the other foot too for the "segregationist" thesis. Recognising "Brazilian Portuguese" as distinct makes it a "real" language rather than a "corrupt" version of the Lisbon dialect, and this, in turn, empowers Brazilians who are not educated to the Lisbon standard. We had this same fight in Canada thirty years ago with French. Noah Webster did the same thing in the US 200 years ago, and for explicitly political reasons. He even wanted to stop calling American English "English", and wanted the US to declare its official language to be "American." Brazilian Portuguese as a social construct would never have left the linguistics community had it not responded to some quite extra-linguistic political needs. So, assigning an explicit political intent to the linguists of the period is not necessarily well founded, even if the origins of this idea could be directly traced to them.