The analysis of communities and pages started with a stage of tracking, which consisted of a procedure to detect communities and pages of groups invested in language and identity problems in Chile, assisted by internal navigation tools of Facebook. It took place in four moments: August 2018, July 2019, November 2019, and May 2020. The process made use of keywords and temporary filters in the Facebook interface of browsers such as Mozilla Firefox and Google Chrome. As described above, the time filters correspond to the dates of every month of the presidential elections of 2009, 2013 and 2017, plus their previous and following months. These temporal frames were chosen due to an assumption of a greater amount of linguistic and political debate within their boundaries. However, given the reticular nature of Facebook content and the hypertextual reference systems that link digital documents (press articles; short texts and comments; images such as memes, graphics, videos, etc.), it was necessary to consider the temporality of hypertextual production as mobile, since it is not possible to establish limits of formal synchrony in the traditional sense. If a digital document (for example, an image and its comments) is posted on a certain date, it can be reposted or made accessible again at later dates, either because the person who produced the document did so voluntarily, or by a suggestion of the Facebook algorithm, which, among other actions, regularly remembers important events and celebrations, as well as texts associated with them.

The keywords searched in the first stages of analysis were: gramática, ortografía, lengua, idioma, diccionario, lingüística, lengua oficial, idioma oficial, oficialización, castellano, español, español de Chile. In the second round were also added lenguaje inclusivo and lenguaje feminista. In the last procedure, chilenismo, and diccionario de chilenismos were also searched.

Since the reach of a Facebook community or page is potentially global, to recognise its correspondence to the Chilean national sphere (that is, to the spatial-political scope of the object of this study), it was necessary to register the locations from where their administrators were operating. These data are available on every Facebook page or community in the “information” section. For the present work, only groups whose managers were entirely (or mostly) within the limits of Chile were selected, thus excluding groups and communities administered from other territories.

The objective of the keyword tracking stage was: 1) to detect groups and Facebook pages focused on language problems in Chile (not only Spanish but also foreign and indigenous languages); 2) appreciate the relative impact of these groups, according to the quantity (and quality) of participants, content and comments; 3) select groups susceptible to critical reading analysis, considering their level of impact.

The second stage of analysis consists of critical reading and includes the examination and discussion of some language ideologies found in the corpus. Particular attention is paid to ideologies described by previous literature, for example, authenticity and anonymity (Woolard, 2007), and pride and profit (Heller and Duchêne, 2012). At first, it was assumed that these categories of analysis in the configuration of Spanish in Chile were highly relevant, which, as discussed below, would not completely occur.

According to the methodology utilised here, the flexibility of access to data and modes of analysis is fundamental in both stages. As Blommaert (2014) suggests, methods leading to the description of complexity require the establishment of new notions of context that contemplate possibilities of meaning as broad as possible, for which it is necessary to