Social Networks: A Source of Lexical Innovation and Creativity in Contemporary Peninsular Spanish

Abstract

There is no doubt that the Internet, where English is ubiquitous, has revolutionized our way of life. Socially, it has opened frontiers to such an extent that nowadays human beings can be permanently connected, no matter the distance between them, in virtual encounters where social networks play a crucial role. Linguistically, on the other hand, it has created a new global language which combines properties of written and oral speech. The Internet’s lexical level, in particular, is described as extremely innovative, creative and playful since it is full of neologisms, many of which are Anglicisms, coined to name the new realities constantly brought along with the evolution of the digital world. In order to demonstrate that social networks are indeed the source of a wide array of creative and playful neological Anglicisms in Peninsular Spanish, we have carried out a corpus-based analysis of the names of five current social and interactive forms of Internet communication in two contemporary Spanish corpora; specifically, that of the indirect social network generically known as the blog and those of the direct social networks Facebook, Instagram, Twitter and WhatsApp. Our study focuses, specifically, on four facets of these neological Anglicisms: (i) their date of introduction into the language; (ii) their frequency of occurrence; (iii) their diverse spellings; and (iv) finally, the different word-formation processes they enter.

1. Introduction

There are two unquestionable facts in our present globalized world and era that prove that Crystal’s (2001) hypothesis about the socio-linguistic revolution brought along with the birth of the Internet holds true. The first one is the development of new means of digital communication, in which the different social networks play a crucial role (Manovich 2001; Vaqueiro Romero 2012; Vila Ponte 2018). The second one is the impact of English, as the unanimously recognized lingua franca of the World Wide Web, on the other world languages (De Mooij 1994; Edwards 1994; Beltrán Llevador 2000; Cronin 2003; Yus 2003; Hjarvard 2004; Montes Fernández 2006; Danbolt Drange 2007; Kowner and Rosenhouse 2008; Balteiro 2011, 2012; Guerrero Ramos 2013; Vettorel 2014; Schmidt and Diemer 2015; García Andreva 2017).

Due to the interrelationship between these two facts, it is not surprising that social networks have become nowadays a very productive source of lexical innovation and creativity in English and in other (world) languages (Smyk-Bhattacharjee 2009; Balteiro 2012; Vila Ponte 2018). Through an exhaustive corpus-based study of the names of five well-known networking services—blog, Facebook, Instagram, Twitter, and WhatsApp—in Peninsular Spanish, this paper analyzes the neological Anglicisms derived from the names of the aforementioned social networking services, paying special attention to the following topics: (i) their date of introduction into the language; (ii) their frequency of occurrence; (iii) the diverse spellings they exhibit; (iv) and, finally, the different word-formation processes they undergo in this particular variety of Spanish.

In addition to this introduction, the present paper contains four other sections. Section 2 addresses the New Age phenomenon of globalization, as well as the important role the contemporary mass media plays in it, so as to understand the privileged sociolinguistic status of English over the rest of the world languages at the present time. Section 3 accounts, in general terms, for the linguistic changes brought about by electronically mediated communication, in which special emphasis is put onto the new vocabulary derived from the names of social networks. Section 4 explains the methodology underlying our corpus-based study and presents its most important findings. Finally, the concluding section gathers the most significant issues raised in our analysis and justifies the neological status of the Anglicisms analyzed.

2. English as the Twenty-First Century Lingua Franca

As Geertz (1986, p. 121) remarks, we are living “in the midst of an enormous collage”, where, as stated in Cronin (2003, p. 169), “peoples, culinary and musical traditions, forms of dress, furnishings, and iconographies are all juxtaposed and thrown together”. There is no doubt, therefore, that we are living in a global world and era (Silverstone 1999, p. 106; Geertz 1986, p. 121; Cronin 2003, p. 169; Williams 2011, p. 28). Nevertheless, understanding what globalization means is not easy because, on the one hand, there is little consensus about what it is (Cronin 2003; MacGillivray 2006), and on the other, as highlighted in Heldt et al. (1999, p. 436), there is “no single coherent theory of globalization and the empirical data generated to assess the impact of global change is limited and contradictory”.

Be that as it may, with these two ideas in mind, globalization can be briefly described, following Robertson (1992, p. 8), as “the compression of the world”. Since “English is in the world and the world is in English”, according to Pennycook (2001, p. 78), it is not surprising that over the last thirty years English has become the unanimously recognized global lingua franca (De Mooij 1994, p. 5; Edwards 1994, p. 41; Posteguillo 2002, p. 18; Hjarvard 2004, p. 76; Montes Fernández 2006, p. 217; Novotná 2007, p. 9; Oncins Martínez 2009, p. 116; Schmidt and Diemer 2015, p. 11). Hence, the term Anglobalization which has been used by Cacchiani (2016, p. 308).

The mass media in the twenty-first century has indeed greatly contributed to the promotion of English as the contemporary international language for several reasons1. First, since it is “assumed to have created a ‘global culture’”, as De Mooij (1994, p. 4) remarks, there is no globalization without media and communications (Rantanen 2005, p. 4; Kelly-Holmes 2005, p. 143). Secondly, the mass media pervades our lives to such a great extent that human beings have become highly dependent on it (Silverstone 1999, pp. 1–2; Gripsrud 2002, p. 3; Jaworski 2007, p. 271). Finally, connecting “us to the world outside”, in Gripsrud’s (2002, p. 4) words, it has such a powerful sociocultural impact on our lives that for Urrutia Cárdenas (2001, p. 12), in particular, it is even greater than that of educational action (De Mooij 1994, p. 121; Jaworski 2007, p. 271).

There is no doubt that the World Wide Web, whose reference language is English (Edwards 1994; Beltrán Llevador 2000; Cronin 2003; Yus 2003; Hjarvard 2004; Danbolt Drange 2007; De la Cruz Cabanillas et al. 2008; Kowner and Rosenhouse 2008; Balteiro 2011, 2012; Guerrero Ramos 2013; Vettorel 2014; Schmidt and Diemer 2015; García Andreva 2017; Vila Ponte 2018), has indeed played a crucial role in the meteoric rise of English in the world languages, since “[i]n recent decades, global communication depends largely on the Internet, as services such as e-mail and applications like Skype and instant-messaging services offered by multiple companies have revolutionized the way in which we communicate” (Reyes and Jubilado 2012, p. 43). Notice, in this regard, that, according to Kowner and Rosenhouse (2008, p. 7), “[a]bout 80% of Internet sites are in English, and most of the programming languages used are based on English”. This figure, though reduced to 53.1% some years later, according to García Andreva (2017, p. 12), still by far surpasses other languages such as Russian (6.4%), German (5.5%), Japanese (5.4%) and Spanish (4.9%).

In sum, due to the “worldliness of English”, as Pennycook (1994, p. 33) calls it, or in similar terms, its privileged linguistic status over the rest of the world languages, as a result of the globalization process we are experiencing in the twenty-first century (Edwards 1994; Brennan 1997; Pennycook 2001; Görlach 2003; Hjarvard 2004; Pulcini et al. 2012), English has changed the status of many of the world languages, among which Hjarvard (2004, p. 76), for example, places different European and Afro-Asian tongues of very diverse linguistic origins—Germanic (German), Romance (French and Spanish), Slavic (Russian) and Semitic (Arabic).

Over the past two or three decades, English has come to occupy a singular position among languages. Previously only one among several dominant European languages, on a par with French or Spanish; it is today a world language, the language people use whenever they wish to communicate with others outside their own linguistic community. English has become the lingua franca of the global network […]. As English has moved toward paramountcy, the status of the other principal languages has changed. Even though they are spoken by more people today than ever before, they have been demoted, degraded in relation to English. Today, French, Spanish, Arabic, German, Russian, etc., more or less have the status of regional languages, national languages that can be used beyond their national frontiers. However, they are losing their currency as languages of international communication, formal and informal—both in political and commercial contexts and in intercultural exchanges, as bridges between people who cross cultural frontiers or who like to enrich their lives with media products from abroad (Hjarvard 2004, p. 76).

The clearest and most direct consequence observed in the world languages, as a result of the influence that English exerts on them, is borrowing. Described as a “natural” and “omnipresent” phenomenon in any language-contact situation by García Morales et al. (2016, p. 33) and Gómez Cápuz (2004, p. 9), respectively, borrowing is to be understood as the exchange of linguistic elements between two linguistic communities or two historical languages. Since “it is at the level of lexis that this phenomenon is greater and most noticeable”, as Oncins Martínez (2009, p. 116) puts it, the most common borrowed elements among languages are lexical items (Görlach 2003, p. 1; Siemund 2008, p. 5; Balteiro 2011, p. 24, 2014, p. 157; Reyes and Jubilado 2012, p. 43; Muñoz-Basols and Salazar 2016, p. 80; García Andreva 2017, p. 13; Sanou 2018, p. 179). Therefore, the frequent attestation of English terms in Spanish, nowadays “undergoing an exponential increase”, according to Núñez Nogueroles (2017, p. 176), cannot be considered an exceptional fact, as verified by their presence, either in their original form or adapted to the orthographic system of the target language, in the following semantic fields: (i) gastronomy (beicon, sándwich); (ii) fashion, clothing and cosmetics (trendy, biquini, rímel); (iii) sports, leisure activities and places (derbi, hobby, resort); (iv) professions (mánager, bróker); (v) politics and economics (líder, marketing); (vi) music and urban tribes (rock, hípster); (vii) cinema, television and shows (tráiler, casting, reality show); (viii) health and fitness (estrés, gym); (ix) housing (hall, lobby); (x) aspects of transport (airbag, sidecar); (xi) mass media and communications (eslogan, spot); and (xii) technology (chat, wifi) (Gómez Cápuz 2004; Balteiro 2011; García Morales et al. 2016; Muñoz-Basols and Salazar 2016; Sanou 2018).

3. The Relationship between Language and the Internet

The quick expansion and the worldwide reach of the Internet in our New Age global era and society have brought along an undeniable sociolinguistic revolution, given that up to now, as Muñoz-Basols and Salazar (2016, p. 87) state, “we have never been more exposed to new cultures and, consequently, new languages”. Socially, it has opened local and international frontiers to such an extent that, as Nolan (1999, p. 1) remarks, “[t]oday, nowhere on earth is truly remote”; thanks to the Web, nowadays human beings can be permanently connected, no matter how far we are from one another, in virtual encounters which Vettorel (2014, p. xxi) describes “as real and meaningful as face-to-face ones”2. Linguistically, it has also made our communication exchanges faster, not to say immediate, and more frequent than not long ago. Given these circumstances, the concept of the “global village”, coined by McLuhan (1964) to refer to the shrinking of the world as a consequence of technological progress, cannot be considered any longer, according to Liu et al. (2011, p. 11), as “an abstract idea but a virtual certainty”, to which we have to be extremely grateful for having allowed us to be close to our loved ones through video calls, video conferences and social networks, thus having helped us emotionally survive the COVID-19 pandemic.

The linguistic exchanges that are produced in the virtual world exhibit, furthermore, several particular features, usually referred to as e-grammar traits, that, as Baym (2010, p. 63) observes, make up “a mixed modality that combines elements of communication practices embodied in conversation and in writing”3. Therefore, it has to be considered a new language, which has been called Netspeak by Crystal (2001, p. 18), the most common distinguishing characteristics of which, widely studied in the literature (Crystal 2001; Baron 2003, 2008; Baym 2010; Androutsopoulos 2011; Herring 2012; Balteiro 2012; Thurlow 2012; Vettorel 2014; Yus 2017a), are the following ones4: (i) orthographically, it is described as being so full of contractions, abbreviations, acronyms, emoticons and emojis, repeated letters, capital letters with connotative meanings, punctuation marks with extremely creative uses, unconventional spellings and self-corrections that, according to Baron (2008, pp. 45–46), it should be considered the “written version of casual speech”; (ii) its most remarkable morphological and lexical traits are creativity and innovation, since it is packed with a huge number of new words formed by means of different word-formation processes, among which affixation, blending, acronymy, conversion and semantic shifts seem to be the most common ones; and (iii) finally, its syntax, filled with frequent omissions and incomplete clauses, as well as with colloquial constructions, is considered, on the one hand, “telegraphic and fragmented” (Herring 2012, p. 5) and, on the other, informal.

Returning to the nature of its lexis, which is our main aim, we have to be aware, first of all, of the incessant terminological creation of new vocabulary, such as, for instance, cookies, tablet, pixels, spam and hacker, which, according to Crystal (2001, p. 67), is motivated by the urgent need to name developments and experiences unknown to date.

Internet users are continually searching for vocabulary to describe their experiences, to capture the character of the electronic world, and to overcome the communicative limitations of its technology. The rate at which they have been coining new terms and introducing playful variations into established ones has no parallel in contemporary language use (Crystal 2001, p. 67).

Though English in their origin, as a direct result of “the hegemony of English in the fastest-growing areas of technological development”, in Cronin’s words (Cronin 2003, p. 121), this innovative and creative lexicon has entered many other languages, among which Spanish is no exception, thus constituting one of the semantic fields more prone to the adoption of Anglicisms (Beltrán Llevador 2000, p. 114; Yus 2003, p. 314; Rodríguez Díaz 2011, p. 78; Reyes and Jubilado 2012, p. 43; Fernández Ulloa 2014, p. xii; Giménez Folqués 2014, p. 45; Muñoz-Basols and Salazar 2016, p. 82; García Andreva 2017, p. 25).

One sphere where Spanish Anglicisms are extremely common is that of information technology, where one finds borrowings like browser, chat, cookie, email, Internet and mail, as well as loan translations like cortafuegos ‘firewall,’ página web ‘webpage,’ ratón ‘mouse’ and ventana ‘window’ […] (Muñoz-Basols and Salazar 2016, p. 86).

This is so because in the other world languages there are not any native terms to refer to these recently created technological inventions, so that, when faced with them, as Guerrero Ramos (2013, p. 118) points out, the language has only one way out: to incorporate a lexical element in its system. Of the three alternatives provided by this scholar, Spanish, not being the international language of science and technology (García Morales et al. 2016, p. 8), has to resort to English since it is the language where these new entities have originally been given a name5. This is, in fact, the most favored option, as the following quotations show:

“A frequently heard answer to this question is that speakers perceive something in the contact language that they feel is missing in their own language. In short, language A lacks what language B has and speakers replicate items to compensate for the perceived gap” (Siemund 2008, p. 10).

“Where there are no native terms for technology and the online-oriented terms/concepts they convey, the foreign terms quickly take root and grow locally” (Dovchin 2016, p. 105).

As their reason of being is clearly due to a “need-feeling motive”, in Anttila’s (1989, p. 155) words, they are necessary and inevitable Anglicisms which, in opposition to gratuitous or superfluous Anglicisms, display a referential function (Alfaro 1948, p. 118; Edwards 1994, p. 76; Urrutia Cárdenas 2001, p. 13; Rodríguez González 2008, pp. 265–66; Balteiro 2011, p. 27; Guerrero Ramos 2013, p. 122; Schmidt and Diemer 2015, p. 15; Vega Moreno and Llopart Saumell 2017, pp. 1441–42)6. Therefore, they are considered, more often than not, a conscious linguistic and cultural enrichment even by the most purist position7. Their referential function confirms, on the other hand, Whitney’s (1875, p. 100) hierarchy of borrowability, later verified in Van Hout and Muysken (1994, p. 42), Kowner and Rosenhouse (2008, p. 12), Pulcini et al. (2012, p. 9) and MacKenzie (2012, p. 31), among other studies, which states, first, that content words are more easily borrowed than function words, and second, that, among the former, nouns are adopted first because, as Van Hout and Muysken (1994, p. 42) observe, “reference is established primarily through [them]”.

4. Lexical Innovation and Creativity in Contemporary Peninsular Spanish through the Names of Social Networks

Within the field of neologisms, those that refer to and are used in social networks occupy a very privileged position, because they are nowadays, as Vettorel (2014, p. 6) remarks, “the preferred means to keep in touch with friends”. Since the presence of English is omnipresent in social networks, we completely agree with Reyes and Jubilado (2012, p. 43) when they state that “[i]n the virtual world, English exerts more of its influence among its speakers as seen in social media such as Facebook, blogs, and other chat avenues and websites”.

4.1. Methodology

As we firmly believe, in the same line as Lorenzo (1996, p. 30), Oncins Martínez (2009), Rodríguez González (2003, p. 574) and Pulcini et al. (2012, p. 18), that “digitized corpora […] have become an indispensable tool for research on English loanwords and their compilation” (Oncins Martínez 2009, p. 116), in the present paper we want to demonstrate through a corpus-based study that social networks are the basic origin of a wide array of neological Anglicisms in contemporary Peninsular Spanish.

In order to reach our aim, we have carried out an analysis of the names of five well-known social networks—blogs, Facebook, Instagram, Twitter and WhatsApp—in two different contemporary Spanish corpora—the last version (beta 0.93 version) of the Spanish Royal Academy Corpus del español del siglo XXI, commonly known as CORPES XXI (2021), and Davies’s (2016) Corpus del español—whose main general features are summarized in

Table 1. General features of the CORPES XXI corpus (Real Academia Española 2021) and Davies’ (2016) Corpus del español.

	Time Span	Total Number of Words	Total Number of Words in Peninsular Spanish	Registers
CORPES XXI	2001–2020	+333 million words	+99.9 million words	Oral/written
Corpus del español	2013–2014	+2 billion words	+459 million words	Websites and blogs

:

Although the names of the social networks chosen for our study may seem random and arbitrary a priori, due to the status of the term blog as a common noun with a generic reference and that of Facebook, Instagram, Twitter and WhatsApp, in contrast, as proper nouns with specific references, there are two main reasons that justify their selection. They all are “forms of Internet communication and networking that clearly emphasize sociality, interactivity and mutuality of information within the generic label of Web 2.0” (Yus 2011, p. 94); hence, their inclusion within what Yus (2017b, pp. 79–80), following Wang et al. (2011), calls “phatic technologies”. Moreover, blogs, which are classified, together with digital forums, as indirect social networking services in Urueña et al. (2011, p. 16), have to be considered, as these researchers state, as the origin of the twenty-first century direct social networks, among which Facebook, Instagram, Twitter and WhatsApp are to be located (Urueña et al. 2011, pp. 13–15). The choice of these four latter social networks, instead of others, is motivated, for its part, by the fact that they are, according to Pato (2020), the most frequently used ones by Spaniards, after YouTube (89%)8: WhatsApp (86%), Facebook (79%), Instagram (65%) and Twitter (53%). Furthermore, we believe that their nature as common or proper nouns may lead to different results in relation to their spelling and the word-formation processes they enter that are worth studying.

We have chosen these two corpora, rather than others, first, because the variety of registers they comprise will show whether the telematic vocabulary recently created around the names of the aforementioned social networks is restricted just to the language used on the Web or if, in contrast, it goes beyond that and has penetrated other linguistic registers. Notice that, whereas Davies’s (2016) Corpus del español is exclusively devoted to the Internet, the CORPES XXI corpus includes oral (10%) and written (90%) registers, among which the material extracted from the Internet only constitutes 7.5% of the total. The remaining written registers in the CORPES XXI corpus are classified as books (40%), press (40%) and miscellaneous sources (2.5%). Secondly, the temporal period covered in the CORPES XXI corpus, which extends from the beginning of the twenty-first century up to the year 2020, makes it possible to investigate whether the social networks that are our focus are indeed the source of new technological vocabulary in Peninsular Spanish9. Notice here that, with the exception of the blog, a discovery born in 1997, the social networking services analyzed are creations of the first decade of our century: Facebook (2004), Twitter (2006), WhatsApp (2009) and Instagram (2010).

In our view, their recent date of creation explains the absence of the proper nouns Twitter, WhatsApp and Instagram in the Oxford English Dictionary (OED) (2018) and in the Spanish Royal Academy Diccionario de la Lengua Española (DRAE) (2014), which, however, include the terms tweet and tuit: tweet in the OED (2018) both as a verb with transitive and intransitive uses and as a noun and tuit in the Real Academia Española (2014) as a noun, from which the verb tuitear and the adjective or noun tuitero derive. Despite being older, the proper noun Facebook is only acknowledged in English as a transitive and intransitive verb, whereas blog, for its part, appears in the two aforementioned dictionaries: in English both as a noun and an intransitive verb (OED 2018) and in Spanish as a noun from which the adjectival or nominal term bloguero has been formed (Real Academia Española 2014)10.

In order to retrieve the different words created around the five terms under analysis in Peninsular Spanish, we have used the * wild card, a substitute of more than one character in both corpora, to the left and the right of the original English terms (blog, Facebook, Twitter, WhatsApp and Instagram) and of their possible Spanish adaptations (Feisbuq, Feisbuk, Tuiter, Wasap and Guasap). Due to the different interfaces both corpora offer, in the CORPES XXI corpus we have specifically looked for the terms at issue by inserting them in the form box and in Davies’s (2016) corpus by placing them in capital letters in the blank box under the List Section.

In the results obtained from this first search we have found many of the words related to Twitter/Tuiter, commonly used in Spanish, missing, such as twittear/tweetear/tuitear or retwitteo/retweeteo/retuiteo, for instance, which, lacking the English derivational -er morph at the end of the name of the aforementioned social network, clearly show that they do not directly derive from it, but from the name of the messages sent and received through it, namely tweets. As a consequence, in order to have a complete picture of the orthographic and morphological patterns of the words created around them, as well as of those built around the name of the social network itself, our search for the social network Twitter/Tuiter has focused on the English and Spanish nouns that name its messages: that is, *tweet*, *twit* and *tuit*.

The total number of examples obtained from both corpora in this first search has been manually analyzed so as to exclude the following cases: (i) words that are not semantically related to the social networks under study, such as gratuitos and fortuitamente, for instance; (ii) Websites, mail addresses and hashtags, like, for example, www.instagram.com, [email protected] and #artistsoninstagram; (iii) proper names, such as Windows MKTwitter and MZBlog; and (iv) English compounds that, like facebookillers and blogroll, have as one of their components an English term with a Spanish counterpart.

4.2. Findings and Discussion

Once the invalid examples retrieved in our first search have been manually discarded, the number of instances that conform to our corpus amount to a total of 439,209 attestations, the raw and normalized (per million words) frequencies of which are distributed as illustrated in

Table 2. Raw and normalized frequencies of the terms under analysis in the CORPES XXI corpus and Davies’s (2016) Corpus del español.

	CORPES XXI		Corpus del Español
	rf	nf	rf	nf
Blog	6540	65.46	281,656	613.62
Facebook	2745	27.47	56,603	123.31
Feisbuk	1	0.01	83	0.18
Feisbuq	--	--	13	0.02
Tweet	200	2.00	7010	15.27
Twit	2959	29.61	53,644	116.87
Tuit	843	8.43	6876	14.98
WhatsApp	704	7.04	10,281	22.39
Wasap	56	0.56	401	0.87
Guasap	7	0.07	168	0.36
Instagram	616	6.16	7803	17.00

:11

4.2.1. Spelling

The first remarkable finding concerns the spelling of these neological Anglisicms, since the five terms examined and/or the numerous words derived from them have been attested in both corpora with different graphic forms: either as raw Anglicisms, thus maintaining their original English form without any kind of modification, or as adapted Anglicisms, thus being formally adjusted to the Spanish orthographic system12. This result confirms, then, that both graphic forms are appropriate, though, as Rodríguez González (1994) observes, they are clearly dependent on registers.

As shown in Table 2, the frequency of occurrence of the English spelling of the words built around Facebook and WhatsApp surpasses by far in both corpora (27.47/123.31 and 7.04/22.39, respectively) those of the two adapted Spanish orthographic forms looked for: in particular, feisbuq (0.00 and 0.02), feisbuk (0.01 and 0.18), wasap (0.56 and 0.87) and guasap (0.07 and 0.36). The adoption of English occurs, despite being orthographically unacceptable in Spanish, according to Pratt (1980), owing to the geminated letters they possess: the vowel o in the former and the consonant p in the latter. As regards the three terms related to Twitter, the English word twit and its derivatives are the ones that, again, present the highest frequency of occurrence (29.61 and 116.87), far surpassing those created around tweet and tuit. In relation to these two terms, however, no sound conclusions can be drawn based on these two corpora: whereas in Davies’s (2016) Corpus del español their normalized frequencies are practically similar (15.27 and 14.98, respectively), in the CORPES XXI corpus the Spanish term tuit (8.43) seems more recurrent than tweet (2.00). In contrast to that of the social networks already mentioned, the spellings of blog and Instagram do not entail significant problems for native speakers of Spanish, despite having at their end, as the previous terms, a consonant, which is not common in final position in Spanish. Therefore, the only form that they exhibit in both corpora is the English one. Nevertheless, different spellings have been attested for the agentive nouns derived from them, either with the Spanish derivational suffix -ero(s)/-era(s) or its English counterpart -er(s), and also for the verbal forms created around blog, which confirms that not only are compounds and words derived through prefixation subject to this graphic fluctuation, as stated in Vila Ponte (2018, p. 105), but also those with inflectional and derivational suffixes. In particular, two orthographic forms of Instagram have been documented with the English agentive suffix: instagramer(s) (three instances) and instagrammer(s) (six examples)13. As regards the spelling of the verbal forms and the agentive nouns created around blog, the situation is far more complex since four different patterns have been attested: (i) the English spelling—bloggear, blogger(s), bloggero(s), bloggera(s)—in which the different affixes are directly adjoined to the base, which doubles its final-g (17,026 occasions); (ii) the Spanish spelling—bloguear, bloguer, bloquero(s), bloguera(s)— where the affixes at issue attach to the base, now with a single -g, by means of an intermediating -u- (8490 times); and (iii) two hybrid orthographic forms in which the English and Spanish spellings combine with each other: in one of them the affixes are directly added to the base without the final-g repeated—blogear, bloger(s), blogero(s), blogera(s)— found on 722 occasions, and in the other they adhere through a linking -u- to the base, now with the final -g geminated—blogguear, blogguer(s), blogguero(s), blogguera(s)—documented in 137 examples.

In our view, the diverse orthographic patterns attested with these five words are a clear sign of their novelty and, consequently, of their non-standardization, thus verifying that the Internet is, as Baron (2008, p. ix) remarks, “a phenomenon in flux”. Their newness is, moreover, reinforced by the fact that in all the cases analyzed the English spelling is the most favored one, since the maintenance of their original English form is, according to Mateescu (2013, p. 66), conclusive evidence of their recent entrance into the language. Although this finding may seem surprising at first sight, because of the registration dates of the English and Spanish spellings in the CORPES XXI corpus (the former always older than the latter), the English graphic forms documented must indeed be considered part of the novel Spanish vocabulary, because, according to their first attestation, they cannot be considered to have long lives: blog (2002), Facebook (2007), Feisbuk (2013), Tweet (2009), Tuit (2010), Twit (2012), WhatsApp (2004), Wasap (2013), Guasap (2015) and Instagram (2011). Notice, furthermore, in this regard that accommodation of any foreign word to the Spanish orthographic system means, for the Spanish Royal Academy, complete integration into the language (Rodríguez Díaz 2011, p. 28; Mateescu 2013, pp. 66–69); or, in a less extreme situation, according to Muñoz-Basols and Salazar (2016, p. 86), it entails a good way to prevent linguistic external influence on Spanish: “By adapting the original English word to the orthographical conventions of the Spanish language and not allowing it to enter the language with its original spelling *twittear, the RAE is also trying to diminish the impact of this foreign word”14.

4.2.2. Word-Formation Processes

The second noteworthy result concerns the vast number of neological Anglicisms constructed, by means of affixation, compounding and blending, around the names of the five digital terms examined, which, as will be immediately shown, share two particular properties. First, they are English/Spanish hybrid terms, where an Anglicism combines with a native Spanish element, either an affix or a base, which, because of the semantic field of new technologies that they belong to, Munday (2015, p. 62) calls “cyberspanglish”. Second, they are extremely innovative and creative, thus corroborating two of the main traits that, as stated earlier, distinguish the morphological and lexical levels of Internet language (Almela Pérez 1999; Crystal 2001; Balteiro 2012; Reyes and Jubilado 2012; Vettorel 2014).

Affixation

The results arising from our work confirm Vettorel’s (2014, p. 161) statement that “[a]mongst the different ways of creating new words, affixation is described as the most frequent and common”. This finding is not surprising, since this word-formation process is responsible for the creation of two kinds of words: on the one hand, those that, by attaching an inflectional morph to an already existing root, are not considered new words, but different forms of the source word; and on the other hand, those that, resulting from joining a derivational affix and a base, are considered, in opposition, novel words.

Inflectional Morphology

As regards inflectional morphology, it is worth highlighting that, besides their expected plural form, due to their nominal status, the five lexical items under analysis exhibit numerous inflectional verbal forms in our corpus. As usually pointed out in the literature, the pluralization of Anglicisms in Spanish, as well as their gender allocation, are two morphological problems brought along with their entrance into Spanish, and these for diverse reasons (Seco 1972; Lorenzo 1994; Gómez Capuz 1998; Domínguez Mejías 2002; Diéguez Morales 2005; De la Cruz Cabanillas et al. 2008; Rodríguez Díaz 2011; Giménez Folqués 2014; Rodríguez González 2017, 2019).

One of the main contrasts between English and Spanish nouns concerns the grammatical category of gender, since, as is well-known, it is a distinguishing feature only of Spanish nouns. Therefore, when incorporated into Spanish, nominal Anglicisms have to be ascribed, on the basis of several criteria, either the masculine or feminine gender, lacking in English (Aron 1930; Haugen 1950; Weinreich 1953; Deroy 1956; Arndt 1970; Zamora 1975; Barkin 1980; Nord 1983; Corbett 1991; Gómez Capuz 1998; De la Cruz Cabanillas et al. 2008; Rodríguez Díaz 2011; Rodríguez González 2019; Núñez Nogueroles 2017)15. Although the nouns that refer to these five social networks are not an exception in this regard, we do not address their gender allocation in the present paper because, since they all end in a consonant, their masculine or feminine gender is not indicated by the inflectional morphs -o/-a, but through the determiners and adjectival modifiers that can potentially appear in the noun phrases of which they are heads.

As regards their pluralization, several hints have also been set forth in the literature (Seco 1972; Lorenzo 1994; Gómez Capuz 1998; Domínguez Mejías 2002; De la Cruz Cabanillas et al. 2008; Rodríguez Díaz 2011; Giménez Folqués 2014; Diéguez Morales 2005; Rodríguez González 2017). In relation to the nominal Anglicisms which end in a consonant, there seems to be, specifically, a threefold choice: to follow either the English or the Spanish regular patterns of pluralization by means of which the inflectional morphs -s and -es, respectively, attach to the singular term, or to omit the plural morph, which, for Gómez Capuz (1998), is a clear case of “morphemic redistribution”. Although these three pluralization patterns have been attested in our corpus, the most recurrent one turns out to be the English model, found with all the different spellings related to the social networks under analysis and in a higher proportion than the other two plural forms, as can be seen in

Table 3. Number of pluralization patterns attested in our corpus-based analysis.

	English Plural (-s)	Spanish Plural (-es)	Zero Plural
Blog	34,787	2	6
Blogg	13	--	--
Facebook	66	--	--
Feisbuk	1	--	--
Instagram	9	--	--
Tweet	3533	--	--
Tweett	1	--	--
Twit	142		--
Twitt	146	1	--
Tuit	1861		--
Tuitt	2	1	--
WhatsApp	124	--	4
Wasap	79	--	--
Wasapp	2	--	--
Guasap	14	--	--
Guasapp	1	--	--

. The Spanish pluralization pattern, in particular, has been only documented on four occasions and just with blog and twit/tuit, and the omission of the plural morph, for its part, has only been attested with blog and WhatsApp and just in ten instances:

Our results thus confirm those provided in the studies by Diéguez Morales (2005) and De la Cruz Cabanillas et al. (2008), where the English regular pluralization pattern was also more productive by far than the Spanish one. This has led Lorenzo (1994) to postulate the birth of a new plural allomorph in Spanish; specifically, the -s that attaches to singular terms that end in a consonant sound.

Finally, with respect to the vast number of verbal forms related to the names of the five social networking services examined, our study reveals that, in opposition to English, where they are the result of the word-formation process known as conversion, in Spanish they are created by means of affixation. As stated in the literature, this kind of denominal verb tends to adopt the first conjugation as it is the most numerous verbal class and the one with the highest functional use in Spanish (Görlach 2003; Solís 2005; Rodríguez Díaz 2011; Reyes and Jubilado 2012). As expected, therefore, it is the inflectional suffix -ar that has been found attached to these five nouns, though not in a productive way. It has been only attested once with blog and Instagram (blogues and instagrama) and in other instances with two of the spellings related to Twitter: specifically, three different verbal forms with the English form tweet (tweetando, tweeta, tweetalo) and five instances with its Spanish version tuit (tuitaba, tuito, tuita, tuitas, tuite).

Far more numerous are the verbal forms in which the morph -ar has been added to the names of the social networks by means of an epentetic -e-, which serves as a link between both elements, thus verifying Reyes and Jubilado’s (2012, p. 46) findings. As shown below, the highest number of finite and non-finite verbal forms—3302 examples in total—has been found with the three different spellings looked for in relation to Twitter: specifically, 2224 cases with tuit, 950 with twit and 128 with tweet. Although in the first two cases the final -t appears both doubled and not doubled, with the Spanish form tuit more verbal forms have been documented with one single -t; however, with its English counterpart twit the situation has been just the opposite:

Tuit(t)ear (2224 instances): tuitear, tuitea, lotuitea, tuiteales, tuitéalas, tuitéa, tuitéame, tuiteemos, tuiteo, tuiteas, tuiteais, tuiteáis, tuiteara, tuiteáramos, tuitearemos, tuiteo, tuitee, tuiteee, tuitées, tuitees, tuiteabas, tuiteaban, tuiteábamos, tuitearía, tuitearías, tuitearían, tuiteen, tuiteé, tuitée, tuiteó, tuiteaste, tuitean, tuitearé, tuitearán, tuitearon, tuitearan, tuitease, tuiteará, tuiteamos, tuiteaba, tuitealo, tuitéalo, tuitearlo, tuitearlos, tuitearme, tuitearla, tuitearlas, tuiteando, tuitéando, tuiteándolos, tuiteadlos, tuiteado, tuitedos, tuiteada, tuiteadas, retuitear, retuiteé, retuiteeis, retuiteemos, retuitearnos, retuitearás, retuitearé, retuitearía, retuiteando, retuiteándo, retuiteándolo, retuiteándolos, retuiteándose, retuitees, retuitee, retuitea, rtuitean, retuiteen, re-tuitea, retuiteáis, retuiteas, retuitean, re-tuitean, re-tuiteamos, retuiteo, retuiteó, retuiteaste, retuiteé, retuitearon, retuitean, retuiteara, retuiteamos, retuiteaba, retuiteaban, retuiteabas, retuiteará, retuitearán, retuitéale, retuitéame, retuitéala, retuitéalo, retuitearlo, retuiteadlo, retuitearlos, retuiteado, retuiteados, retuiteada, tuittear, tuittea, tuitteando;

Twit(t)ear (950 instances): twittear, twittearlo, twittearlos, twittead, twittearon, twittees, twitteen, twitteara, twitteaba, twitteabas, twitteaban, twitteando, twitteáis, twittearla, twitteadlo, twitteado, twitteada, twitteadas, twittea, twittearán, twitteas, twitteamos, twitteo, twittee, twittean, twittéalo, twittealo, twittearía, twittearlas, twitteó, twitteé, retwittear, re-twittear, retwittearlo, retwittearlos, retwitteandote, retwitteado, retwitteada, retwitteados, retwittea, retwitteas, retwittean, retwittee, retwitteen, retwitteo, retwittearé, retwittearán, twitear, twitearlo, twitearlos, twiteando, twiteado, twiteados, twiteada, twiteo, twitees, twitea, twiteas, twitean, twiteen, twitee, twiteaba, twiteó, twiteé, twiteaste, twitéame, twiteara, retwitear, re-twitear, retwitearte, retwitearlo, retwiteo, retwiteó, retwitea, retwitee, retwiteen, retwiteaste, retwitease, retwiteando, retwitean, retwiteada, retwiteadas, retwiteado, retwitearas, retwiteaba;

Tweetear (128 instances): tweetear, tweetearlo, tweetearon, tweetees, tweeteo, tweetea, tweeteas, tweetean, tweeteado, tweeteados, tweeteada, tweeteadas, tweeteando, tweeteará, tweeteaba, retweeteando, retweetearlo, retweeteo, retweetee, retweeteé, retweeetas, retweeteados, retweeteadas, retweetearos, retweetearlos, retweeteen, retweeteó, re-tweeteas, retweetea, retweetean, retweeteado, retweeteados, retweetear.

The term blog occupies the second place, having yielded a total number of 620 attestations, which also exhibit different spellings. Here again, the Spanish orthography, found in 511 examples, is the most productive one, followed, though at a great distance, by the English spelling, documented, for its part, in 82 instances. Finally, the two hybrid English/Spanish graphic forms have been attested, but with lower frequencies of occurrence; in particular, 21 examples have been retrieved for the verb blogear and only six instances for its counterpart blogguear:

Bloguear (511 examples): bloguear, blogueando, bloguea, blogueo, bloguean, blogueamos, blogueado, bloguearía, bloguearé;

Bloggear (82 examples): bloggear, bloggeando, bloggeara, bloggeamos;

Blogear (21 examples): blogear, blogeando, blogeado;

Blogguear (6 examples): blogguear, bloggueando, bloggueaba.

Despite also presenting English and Spanish spellings, the verbal forms built around the term WhatsApp appear in the third place but with a much smaller number of attestations: 89 in total. In contrast to those created around Twitter and blog, here the hybrid English/Spanish spelling wasapear is the most recurrent one, presenting a fairly similar level of frequency to that of the English original form whatsappear (36 vs. 32 attestations). The most-adapted Spanish orthography guasapear is, for its part, the least common graphic form of the three ones searched for in relation to this social network (21 instances):

Wasapear (36 cases): wasapear, wasapearse, wasapeando, wasapeo, wasapea, wasapeamos, wasapees, wasapeaba, wasapeaban;

Whatsappear (32 cases): whatsappear, whatsappearse, whatsappearla, whatsappeando, whatsappea, whatsappean, whatsappeo, whatsappeamos, whatsaapeabas;

Guasapear (21 cases): guasapear, guasapeando, guasapeado, guasapeo, guasapean, guasapeas, guasapeaba.

The verbal forms created around the social network Instagram are much less numerous than the ones previously discussed, having been documented on only 22 occasions. In our view, there are two reasons that account for their low frequency of occurrence. First, this is because only the Spanish spelling, in which the final -m- of the base is not doubled, has been attested, and second, because, together with the three nonfinite verbal forms, only three different finite verbal forms have been documented in our analysis. This second fact also explains the reduced number of occurrences—29 attestations in total—that the search with Facebook has yielded. Nevertheless, this particular finding seems to us somehow unexpected for two reasons: on the one hand, because, after the term blog, Facebook is the oldest social network that we have analyzed; and, on the other, because it has been attested with three different spellings:

Instagramear (22 cases): instagramear, instagrameo, instagrameen, instagramean, instagrameando, instagrameada, instagrameadas, reinstagrameado;

Facebookear (16 cases): facebookear, facebookeando, facebookeáis, facebookea, facebookeo;

Feisbuquear (10 cases): feisbuquear, feisbuqueando, feisbuqueado, feisbuqueada, feisbuqueo, feisbuquea;

Feisbukear (3 cases): feisbukear, refeisbukear.

It is worth noticing at this point, moreover, that, the epentetic -e- has been substituted in our corpus, though not very frequently, by an -i-. Specifically, it has been attested only in the infinitive and the gerund forms built around Twitter and just on three occasions, with one case for each of the different spellings examined: specifically, tuitiar, twittiando and tweetiarlo.

Derivational Morphology

The new terms created from the social networking services under analysis through the word-formation process called derivation should be included within Haugen’s (1950) “hybrid derivatives”. In these derivations, an autochthonous affix—a prefix, a suffix or both—is attached to a foreign root. In this regard, our study has demonstrated, in opposition to García Andreva (2017, p. 12), who considers that hybrid derivatives are very low in number, that derivation is a very productive source of creating new technological vocabulary in Spanish. Specifically, we have found twenty-one Spanish prefixes and twenty-eight suffixes attached to the names of the social networks that are of interest.

Concerning prefixation, re-, which means “repetition”, has been undoubtedly attested as the most productive prefix in our corpus, documented on 130 occasions with four of the five social networking services analyzed: blog, Twitter, Facebook and Instagram. In particular, the terms related to Twitter, no matter their spelling, are the ones which have yielded, by far, more forms; specifically, 57 instances with the Spanish orthographic form tuit and 45 and 28, respectively, with their English versions twit and tweet:

*Tuit* (57 instances): retuit, retuits, re-tuits, rettuit, rettuits, retuitear, retuiteé, retuiteeis, retuiteemos, retuitearnos, retuitearás, retuitearé, retuitearía, retuiteando, retuitéando, retuiteándole, retuiteándolo, retuiteándolos, retuiteándose, retuitees, retuitee, retuitea, rtuitean, rtuiteen, re-tuitea, retuiteáis, retuiteas, retuitean, re-tuitean, re-tuiteamos, retuiteo, retuiteó, retuiteaste, retuiteé, retuitearon, retuitearan, retuiteara, retuiteamos, retuiteaba, retuiteaban, retuiteabas, retuiteará, retuitearán, retuitéale, retuitéame, retuitéala, retuitéalo, retuitearlo, retuiteadlo, retuitearlos, retuitedao, retuiteado, retuitados, retuiteados, retuiteada, retuiteadas;

*Twit* (45 examples): re-twit, retwit, retwits, retwitear, re-twitear, retwitearte, retwitearlo, retwiteo, retwiteó, retwitea, retwitee, retwiteen, retwiteaste, retwitease, retwiteando, retwitean, retwiteada, retwiteadas, retwiteado, retwiteados, retwitearas, retwiteaba, retwitt, retwitts, retwitter, retwitters, retwittear, re-twittear, retwittearlo, retwittearlos, retwitteándote, retwitteado, retwitteada, retwitteados, retwittea, retwitteas, retwittean, retwittee, retwitteen, retwitteo, retwittearé, retwittearán, retwittar, retwittado, retwitteadlo;

*Tweet* (28 instances): retweet, rtweet, re-tweet, retweets, re-tweets, retweeters, retweeteando, retweetearlo, retweeteo, retweetee, retweeteé, retweeteas, retweeteados, retweeteadas, retweeteable, retweeteables, retweetearos, retweetearlos, retweeteen, retweeteó, re-tweeteas, retweeteas, retweetea, retweetean, retweeteado, retweeteados, retweeteadores, retweetear.

With the remaining social networks, however, the words with the aforementioned prefix in our corpus are fewer in number: specifically, only two with Instagram (re-instagram, re-instagrameado) and one with Facebook and blog (refeisbukear, reblogueado).

Less numerous are the cases found with the prefixes pre- (five cases), auto- (four instances), tele- and audio- (three cases) and geo- and mini- (two examples), despite having been also attested with more than one social network: pre-Facebook, pretwitter, pre-tuiter, pre-instagram, prewhatsapp, autotweets, auto-tweet, autotuits, autoblog, teletweeting, teletuiterizate, teleblogdiario, audioblog, audioblogs, audiowasaps, geotweets, geo-blog, miniblog, and mini-facebook. The least productive are, however, the ones that follow, given that, no matter the number of examples retrieved with them, they have been found only with one of the social networks under analysis. With the different spellings searched for in relation to Twitter we have found, specifically, eleven different prefixes: (i) mega-: megatweet; (ii) giga-: giga-tweet; (iii) poli-: politweets; (iv) pro-: protwitter; (v) des-: desintituitearme; (vi) in-: desintituitearme; (vii) metro-: metrotwit; (viii) biblio-: bibliotuits; (ix) requete-: requeteretuiteadas; (x) tecno-: tecnotwitter; (xi) and súper-: super-tweeter, supertweeter. Blog, for its part, has been found twice prefixed with foto- and nano- and on three and five occasions, respectively, with micro- and video-: fotoblog, fotoblogs, nanoblogs, nanoblogging, microblog, microblogs, microblogging, videoblog, videoblogs, video-blog, videobloggers, and videoblogging. Finally, Facebook has only derived two words with the prefix anti-: antifacebook, anti-Facebook.

On the other hand, with respect to suffixation, the most recurrent suffixes in our corpus are nominal in nature. Of the twenty-one nominal suffixes identified, the Spanish and the English agentive suffixes -ero(s)/-era(s) and -er(s)/-or(s) are the ones with the highest frequencies of occurrence, having been registered with the five social networks examined on 50 and 35 occasions, respectively. They coexist, furthermore, with the Spanish suffix -ador(es)/-adora(s), which is also agentive in nature, but much less frequent, having been found just five times and only with the Spanish spelling related to Twitter:

-ero(s)/-era(s) (50 cases): facebookero, facebookera, facebookeros, facebookeras, feisbuquero, feisbuquera, instagramero, instagrameros, instagrameras, tweetero, tweetro, tweetera, twettero, tweetteros, twitero, twiteros, twitera, twiteras, twittero, twitteros, twittera, twitteras, tuitero, tuiteros, tuitera, tuiteras, retuiteros, blogero, blogeros, blogera, blogeras, bloguero, blogueros, bloguera, blogueras, bloggero, bloggeros, bloggera, bloggeras, blogguero, blogueros, blogguera, bloggueras, blogueiros, whatsappero, whatsapperos, guasapero, guasaperos, wasapero, wasaperas;

-er(s)/-or(s) (35 instances): facebookers, instagramer, instagramers, instagrammers, instagrammers, tweeter, tweeters, twiter, twiters, twitter, twitwer, twittwer, twittter, twitters, twitteres, twiteer, twitterer, twitterers, twitor, tuiter, tuitter, tuiters, tuiteer, bloger, blogers, bloguer, bloguers, blogger, bloggers, bloggeres, blogguer, blogguers;

-(a)dor(es)/-(a)dor(as) (5 cases): tuiteador, tuiteadora, tuiteadores, retuiteador, retuiteadora.

Less productive are the following seven nominal suffixes, documented, for their part, not with all the social networks under study, but with at least one: (i) -eo (ten cases): twiteos, twitteos, retwitteos, retwiteos, retuiteos, re-tuiteos, whatsappeo, whatsappeos, wasapeo, guasapeo; (ii) -ano(s)/-ana(s) (nine cases): facebookiano, facebookiana, facebookianos, facebookianas, twitteriana, twitterianos, twiteana; tuiterianos, whatsappiano; (iii) -ría (five cases): twitoría, retuitería, tutería, tuiterías, bloguería; (iv) -ción (three cases): facebookización, instagramización, tuiterización; and (v) -(r)rea, (vi) -landia, (vii) and -metro, attested twice: tuitorrea, blogorreico, facebooklandia, twitterlandia, tuitómetro and blogómetro.

Finally, the eleven remaining nominal suffixes identified have been found only with one of the social networks analyzed:

*Twitter*: (i) -ario and (ii) -azo (two cases): twitario, tuitario, twittazo, tuitazo; (iii) -ío, (iv) -onia and (v) -grafía (one case): twitterío; twitteronia, twittergrafía;

*Blog*: (i) -ismo (four cases): blogerismo, bloguerismo, bloggismo, bloguismo; (ii) -logo (three cases): blogólogo, blogólogos, blogológico; (iii) -ita, (iv) -logía and (v) -(c)ito (one case): blogalita; blogología; blogcito;

*Whatsapp*: (i) -itis (one instance): whatsappitis.

Apart from these nominal suffixes, four different adjectival suffixes, two distinct verbal suffixes and one single adverbial suffix have been registered in our corpus, either with just one or with more than one of the social networks under study, but never with all of them:

Adjectival suffixes: (i) -able(s) (nine cases): instagramable, instagrameables, retweeteable, retweeteables, twitteable, tuiteable, tuiteables, retuiteable, retuiteables; (ii) -ico(s)/-ica(s) and (iii) -il(es) (five cases): instagramicas, tuitérica, blogorreico, blogológico, blogológica, facebookil, twitteril, tuiteriles, blogueril, wasapiles; and (iv) -ensis (one case): facebookensis;

Verbal suffixes: (i) -izar (seven instances): facebookizar; instagramizar, instagramizando, instagramizada, instagramizadas; tuiterización, teletuiterizate; (ii) -(i)ficar (one instance): tuitificarlo;

Adverbial suffix: (i) -mente (two examples): twitteramente, tuiteramente.

Our results thus confirm Vettorel’s (2014, p. 131) statement that “innovations have indeed been shown not to be erratic, but to follow well attested language change processes”.

Compounding

The word-formation process known as compounding, more frequent in English than in Spanish (Görlach 2003, p. 78; Piera 1995, p. 302), has also been attested in our corpus as the source of a wide array of new vocabulary related to the digital world. As illustrated in

Table 4. Compounds attested in our corpus-based analysis.

*Blog* (45 compounds)	blogacción, eventoblog, cumpleblog, blogoliteraria, escritura-blog, blogosfera, blogósfera, blogosférico, blogosférica, blogosféricos, blogosféricas, blogsfera, blogoesfera, blogesfera, blogosferas, bloggosfera, blogocosa, blogonovela, blogonovelas, blogonovelista, blognovelas, blogocultura, blogorreportaje, teleblogdiario, diarioblog, iniciablog, blogalaxia, bloglector, blog-lector, bloglectores, bloguero-cotilla, blogoeventos, blogestudio, bloguionistas, blogoplancton, blogocéntricas, blogmaníacos, blogdecine, blogmaster, weblog, weblogs, weblogger, webloggers, blogger-blogger, egoblogger.
*Tuit* (40 compounds)	sextuitero, sextuiteros, sextuitera, mamatuitera, mamatuiteras, tuitósfera, tuitosfera, tuitesfera, tuitagitadores, tuitdebates, tuitcomedia, tuit-amigo, tuitrafico, papatuitea, tuit-zarandaja, tuitbiografía, tuithistorias, tuiterbicho, bloguero-tuiteros, blogueros-tuiteros, chatuitero, cineclubtuitero, blogtuitesfera, tuitorienta, tuitorientación, tuitorientador, tuitorientando, tuitmanifestación, tuiteractúa, tuitexperimento, tuitredada, tuitnovela, tuitnovelas, tuiter-adicto, blog-tuit, blog-tuitee, tuiteatrera, tvtuit, tuit-predicadores.
*Twit* (39 compounds)	twittercompartir, twitlogo, twitidea, twitencuentro, twitterbronca, twitternovela, twitterdinámica, twitterperiodismo, twitterhéroes, twittesfera, twitosfera, twittosfera, twitteresfera, twittersfera, twitteresférico, twitterdependiente, twitter-facebook, facebooktwitter, twitterdependientes, twitter-adicto, twitterverborrea, blog-twitter, blogueros-twitteros, twittermanías, twitterevento, twittervisión, twittervista, twittertulia, censuratwitter, listastwitter, twitter-revolución, twitteradicto, twitteradictos, informetwitter, todotwitter, twit-herramientas, twitherramientas, twitt-disgustos, sabiduríatwittera.
*Tweet* (16 compounds)	twittercompartir, twitlogo, twitidea, twitencuentro, twitterbronca, twitternovela, twitterdinámica, twitterperiodismo, twitterhéroes, twittesfera, twitosfera, twittosfera, twitteresfera, twittersfera, twitteresférico, twitterdependiente, twitter- facebook, facebooktwitter, twitterdependientes, twitter-adicto, twitterverborrea, blog-twitter, blogueros-twitteros, twittermanías, twitterevento, twittervisión, twittervista, twittertulia, censuratwitter, listastwitter, twitter-revolución, twitteradicto, twitteradictos, informetwitter, todotwitter, twit-herramientas, twitherramientas, twitt-disgustos, sabiduríatwittera.
*Facebook* (6 compounds)	Facebook-adictos, Facebookcorreo, comentariosfacebook, Tipofacebook, Amigo-Facebook, Fotos-Facebook.
*Instagram* (3 compounds)	instagramego, todoinstagram, foto-instagrameándonos
*Whatsapp* (1 compound)	whatsappfoto.

, blog, the oldest term of the five ones examined, is the one which has yielded more compounds. The second place is occupied by the three different graphic forms associated with Twitter. Of these three, however, the Spanish spelling tuit and its English counterpart twit are the origin of the highest number of compounds, 40 and 39 different examples, respectively, thus virtually doubling the attestations retrieved with its other English version tweet (16 instances). Finally, at a great distance in frequency are the compounds created around Facebook, Instagram and WhatsApp, having been attested only on one, three and six ocassions, respectively:

As these examples show, the names of the social networking services under analysis appear both as the first and the second components in the compounds attested in our corpus. Therefore, because of the prototypical structure of Spanish compounds, where, in contrast to English, the first component functions as the head and the second one, for its part, as the modifier—as illustrated, for instance, in lengua madre and perro policía (Piera 1995; Ulašin 2016)—the English terms in these compounds could be thought a priori to display both functions. However, a careful analysis of the compounds previously listed reveals that their English component only acts as a modifier, no matter its position: mamatuitera, informetwitter, sabiduría-twittera, comentariosfacebook, amigofacebook, blogonovela, bloglector, blogmaníacos, twitteradictos, and whastappfoto. This is an interesting finding because, due to the right-headedness of English compounds, it entails that in those cases in which the name of the social network is placed first, which are far more numerous in our corpus than those where it is the second component, not only the English term, but also the morphological structure prototypical of English compounds (modifier plus head), have been borrowed into Spanish.

Blending

Some of the neological terms attested in our corpus are the result of blending, a playful and extremely creative word-formation process (Cacchiani 2016, p. 306; Beliaeva 2019, p. 3), halfway between clipping and compounding, which, according to Ulašin (2016, p. 163), is marginal in Spanish and, therefore, not so productive as in English. Defined, in general terms, as an “ingenious way of associating the semantics of two or more words in a new morphological object” (Renner 2015, p. 119), it should undoubtedly be considered as an expressive means of language that creates new lexical items described as “attention-catching and thought-provoking” by Beliaeva (2019, p. 2).

Though blending prototypically builds new words, like brunch and motel, for instance, by joining together the initial part of a word that already exists in the language (breakfast and motorway) and the final part of another (lunch and hotel), blends may adopt other forms16. As illustrated in

Table 5. Blends attested in our corpus-based analysis.

	Beginning of Word (15 Instances)	Middle of Word (18 Instances)	End of Word (12 Instances)
*Blog*	blogobate, blogatón, blogfesor, blogfesores, blogodismo	ablogado, biblografía, biblogtecarios	edublog, edublogs, moblog(s), liblog
*Tuit*	tuitiqueta, tuiteratura, tuitkipedia	batuitizado, actuitud, cuentuitos, gratuitud, gratuitdad, instituitos, constuituir, constituituyen, constituituirían, restuitirá, protuituirse	poetuit, poetuits, poetuitero, poetuiteros, edutuitera, cortocirtuitando
*Twit*	twittersodios, twitterfera, twitteratura		edutwitter, edutwitters,
*Tweet*	tweetulares, tweetpatía,	Elquijotweet, Celestweetna, Lacelestweetna, experiment- tweetado
*Facebook*	facebookteracción,
*Instagram*	instagramdicta,

, two of them, specifically, have been attested in our corpus: in one of them, the English name of the social network appears, without being cut, as the first or the second component of the new term, and in the other, called infixed blends (Bauer 2012) or central replacement blends (Beliaeva 2014), it is inserted, also in its complete form, in the middle of an already existing word:

As illustrated above, the names of all the social networks examined, except for WhatsApp, are the sources of these three kinds of blends in Spanish. However, blog and the different graphic forms related to Twitter are the most recurrent ones, perhaps due to their length, appearing in the three positions previously listed. Facebook and Instagram, for their part, have only been documented once each and in both cases in the initial position. After a detailed analysis of the examples given, three important conclusions can be drawn. First, the English name of the social network is never cut, which verifies Beliaeva’s (2019, p. 8) remark that “the constituents of blends have to remain potentially recognisable”, thus making them understood and, as a consequence, successful. Second, when only one of the two components is shortened, it is not always the first one, as facebookteracción, instagramdicta, tweetpatía, twittersodios, tuiteratura and blogfesor, for instance, show. Finally, infixed blends, the most numerous ones in our corpus, should not be considered so peripheral, as Beliaeva (2019, p. 8) observes, at least in the field of new technologies.

4.2.3. Registers

The final remarkable finding derived from our study concerns the frequency of occurrence of the new terms built around the names of the five social networks under analysis in the different registers comprised in the CORPES XXI corpus, where, as illustrated in

Table 6. Raw and normalized frequencies of the terms under analysis in different registers (CORPES XXI corpus).

	WRITTEN				ORAL
	Books	Press	Miscellanous	Web
Blog	2060 (34.12)	1638 (35.14)	17 (25.41)	2785 (568.29)	40 (4.00)
Facebook	910 (15.06)	1303 (27.96)	4 (5.98)	498 (101.61)	30 (3.00)
Feisbuk	1 (0.01)	--	--	--	--
Tweet	17 (0.27)	113 (2.42)	--	70 (14.27)	--
Twit	907 (15.02)	1449 (31.09)	4 (5.98)	566 (115.47)	33 (3.30)
Tuit	377 (6.24)	320 (6.86)	--	131 (26.72)	15 (1.50)
WhatsApp	470 (7.78)	179 (3.83)	--	40 (8.15)	15 (1.50)
Wasap	44 (0.72)	4 (0.08)	--	7 (1.42)	1 (0.10)
Guasap	2 (0.03)	3 (0.06)	--	2 (0.40)	--
Instagram	229 (3.78)	243 (5.20)	2 (2.99)	123 (25.09)	19 (1.90)

, their presence is well-attested:

These results are significant, for they clearly show that the Web is not the only register in which this kind of innovative, playful and creative vocabulary is used. Although a comparison of their normalized frequencies of occurrence verifies our initial hypothesis that stipulates that it is on the Internet that these neological Anglicisms are most recurrent, it also reveals that books and the press are two suitable registers for their use, since their presence is sufficiently documented in them. Thus, in agreement with Guerrero Ramos (2013, p. 126), we can conclude that, as in other specialized languages, once the neologisms born in certain registers are settled, they begin to lose their distinctive status so that they enter other linguistic registers to such an extent that they become “widely used […] in face-to-face contexts as part of daily communicative events”, as Dovchin (2016, p. 105) states, even “polluting traditional writing” (Baron 2008, p. 163). Therefore, given their ubiquity in everyday language (Zirker and Winter-Froemel 2015, p. 1; Cacchiani 2016, p. 320), creativity and wordplay have to be understood, following Carter (2008, p. 430), as linguistic properties “actively possessed by all speakers and listeners” and “not simply the domain of a few creatively gifted individuals”.

5. Conclusions

As Daniels (1985, pp. 18–63), among others, states, there is no doubt that languages are intimately related to the societies and individuals who use them. Language transformation is, thus, inevitable and unstoppable because human beings live in a permanent state of flux. As new realities are constantly becoming part of our daily lives, they need to be named. Therefore, these changes are mainly seen in the lexical repertoires of languages. In this regard, our corpus-based study is one more evident proof of the constant evolution that languages experience, since it demonstrates how the lexicon of Peninsular Spanish has shifted in the twenty-first century with the incorporation of a considerable number of new terms created around the English names of different social networking services; namely blog, Facebook, Twitter, WhatsApp and Instagram. Specifically, the 439,209 analyzed attestations reveal that these names are the sources of a wide array of graphically and morphologically different terms, created by means of affixation, compounding and blending.

Although they have been subject to different classifications—formal neologisms (Vega Moreno and Llopart Saumell 2017, p. 1419), borrowed neologisms (Auger and Rousseau 2003, pp. 79–84) and interlinguistic neologisms (Bermúdez Fernández 1997, pp. 14–15)—their status as neological Anglicisms cannot be questioned since, besides their English origin, which is evident, the terms attested in our study satisfy, as made clear throughout our work, most of the different criteria that define neologisms postulated in the literature (Rey 1976; Cabré Castellví 1993; Estornell Pons 2009; Méndez Santos 2011; Vega Moreno and Llopart Saumell 2017). First, the social networks under analysis, except for blog, which is a digital reality born at the end of the twentieth century (1997), are discoveries of the first decade of the twenty-first century. Therefore, their names, having a short life, fulfill the chronological criterion, which states that neologisms are terms which have recently entered the language. As a consequence, then, they are usually absent in dictionaries, thus confirming the lexicographical parameter that determines neology. Notice in this regard that only the terms blog and tuit are included in the: Real Academia Española (2014). Furthermore, their raw and normalized frequencies of occurrence in both corpora prove that they are recurrent terms in different linguistic registers, thus verifying the neological criterion usually referred to as “usage spread”. Finally, their different spellings and graphic forms are a clear sign of formal instability, which is another criterion that also seems to determine neologicity17. However, their involvement in several word-formation processes, being the base of countless innovative, creative and playful terms, is, according to Guerrero Ramos (2013, p. 126), in contrast, a clear sign that they are beginning to lose their neological status and, therefore, of their emerging integration in the recipient language. Nevertheless, due to the ephemeral nature of the Internet, as a consequence of the constant and rapid changes it undergoes every day, it is interesting to ask, along with Vettorel (2014, p. 80), “whether these new Internet words will […] lead to long-term language”. The only thing we can do is to wait and let time pass in order to find out.