Corps de l’article

1. Building the notion of translation policy

The idea that there is something scholars can call “translation policy” has existed since the very early days of the field known as Translation Studies (Chesterman 2009: 14). In those early days, however, the notion was peripheral to the concerns of most scholars. As Reine Meylaerts (2011a: 163) observed, during the first roughly four decades of the field, the term translation policy meant many different things to different scholars so that it was hardly a useful notion. Fortunately for researchers, this situation has changed over the last decade, and while the notion of translation policy continues to be difficult to define around the edges, it has been developed considerably.

Oscar Diaz Fouces was one of the first scholars to begin writing about the role of translation in matters of language policy and planning (e.g., 1996). He noticed (2001: 121-123) the overall absence in the relevant literature of discussions regarding the role of translation (and its practitioners!) as elements of cultural policy. As Diaz Fouces explored that role, he also argued (2002: 85-56) that matters of linguistic mediation can be distinguished from language policy in and of itself.[1] These are important contributions in that they anchor translation policy as a type of public policy that can be a worthwhile object of study.[2]

In turn, Meylaerts brought the study of translation policy to the forefront, in part through an insightful entry in the Handbook of Translation Studies (2011a). Elsewhere, she began identifying different types of translation policy regimes and argued that “there is no language policy without a translation policy” (Meylaerts 2011b: 744). If the existence of a language regime means that choices have to be made about translation, then translation choices have to be made in official settings (see for example González Núñez 2016a) but also in non-official settings, including by actors in civil society (Tesseur 2017, among others). Meylaerts’ insights explicitly establish translation policy as a consequence or corollary of language policy.

It follows, then, that lessons might be learned from studies on language policy, such as those made by Bernard Spolsky (e.g., 2012). In this sense, I have argued (González Núñez 2016b: 50-56) that a methodologically and conceptually useful approach to the study of translation policy includes observing translation management, translation practice, and translation beliefs. Using this approach, I have identified other types of policies, beyond language policies, that help shape translation policy. These include policies to promote human rights and policies to protect ethnic minorities (e.g., González Núñez 2016b: 62-93). In turn, Hlavac, Gentile, et al. (2018) have identified the promotion of multiculturalism as a public policy where translation emerges in order to serve other primary objectives. One can easily conclude that translation policy is shaped by a number of factors, including things as mundane as whether the policy is the result of “explicit translation management rules” or, conversely, the result of “pragmatic problems solving” practices (González Núñez 2017: 164). In other words, translation policy is about more than language policy alone. Indeed, translation policy stems from the interaction of different public policies that have linguistic elements.

As studies continue to spring forward, new perspectives continue to enrich our understanding of translation policy (for instance the special issue of International Journal of the Sociology of Language edited by Córdoba Serrano and Diaz Fouces in 2018). These perspectives can come not only from the field of Translation Studies but also from economists, political philosophers, legal scholars, etc. (e.g., the collected volume edited by González Núñez and Meylaerts (2017)). Generally speaking, these studies tend to look into the role that the authorities play in shaping translation policy. Such an approach has led to useful insights, and for the foreseeable future, it will probably continue to provide enriching perspectives.

A commonality in all these studies is that translation policies are developed in a context of social multilingualism. In a way, this is to be expected. In order for translation policies to take shape, some part of society, some domain within it, must operate multilingually. On this point, Grin (2017: 160-161) has rigorously argued that social, economic, and political factors – including language policies – contribute in creating social multilingualism, and consequently these determinants help shape translation in such societies. Therefore, all of these determinants interplay in a dynamic way to give shape to translation policy in a given place at a given time.

Additionally, as Hlavac, Gentile, et al. (2018: 63) have observed: “policy formation, at least in countries with a democratic tradition, is a multi-party process with macro-social dynamics that intersect with individual events and protagonists, sometimes in a separate way, sometimes in a cumulative or collective way.” A further consideration is that language itself has been described as a “complex adaptive system” where multiple agents interact over time at different levels in ways that are anything but static (Beckner, Ellis, et al. 2009). Thus, in dealing with language and translation policy, several levels of complexity are at play. This is evident in the many disciplines that language policy crosses paths with, including economics, political theory, law, education, and so forth (see Ricento 2005). Many of these overlaps are also observed in translation policy. In light of this, interdisciplinary research endeavors like the MIME project, which considers together “a wide range of language issues [including translation] that are usually addressed separately,” are helpful in dealing with the complexity of such issues (Grin 2019: 15).

This leads us to concluding that if as scholars we are to consider why translation policy looks the way it does in any given place and time, we can consider a number of factors beyond the pressure exerted by policies about language, human rights, minority protection, or multiculturalism. In this regard, we can consider different processes, dynamics, events, and even individuals who help shape translation policy. In other words, there are insights to be gleaned by looking beyond the official domains themselves.

In that spirit, this article will look beyond the interaction of the policies themselves in order to help further the understanding of how translation policy takes shape. To pursue this objective, it will adopt a descriptive and comparative approach. The paper will consider translation policy in two different settings, namely Bangor, Gwynedd (Wales) and Brownsville, Cameron (Texas). The information presented in this study comes from policy documents, Freedom of Information requests, and interviews. Section 2 will broadly compare and contrast these two regions in terms of history and demography. Section 3 will then outline translation policy in both regions, which will serve to highlight some stark contrasts. Section 4 will provide an analysis of what elements from section 2 can help account for what is found in section 3. Section 5 will conclude this study by arguing that a fundamental difference in how the minority language is viewed in these regions is largely responsible for the different shapes observed in their translation policies.

2. A short comparison between Gwynedd and Cameron counties

To understand how Gwynedd and Cameron counties are different and yet similar, this section will provide a broad overview of each region’s history and demographics. What eventually would be known as Cameron County is located in the southernmost tip of Texas, in an area known today as the Lower Rio Grande Valley. Originally populated by indigenous hunter-gatherers, the region came under Spanish control in the late 1600s. Eventually, native languages would disappear from the area as indigenous people were either killed or absorbed by the conquering cultures.

In 1810, Mexico began a war of independence that would lead to the area going from Spanish to Mexican control. Mexican Texas was a Spanish-speaking territory, but it showed some tolerance to English speakers as these started coming into the northeastern part of the state during the early 1800s (Blanton 2004: 13-16). Tensions between Anglo settlers and Mexican authorities exploded into an all-out war that led to the independence of Texas in 1836 (Balestra, Martínez, et al. 2008: 17). This brought about a switch in language policy: English became the official language, and Spanish was tolerated, at least initially (Blanton 2004: 16). This tolerance was limited. General attitudes from Anglos toward Mexicans and the Spanish language were negative, so in the Republic of Texas the Spanish language was seen with disdain (Martínez 2005: 242-244).

The United States (US) likewise “exerted pressure on the Spanish borderlands, both from the east and from the north, encouraged by a new expansionist dream” (Balestra, Martínez, et al. 2008: 14). In 1845, the US federal government and the Republic of Texas both approved Texas’ annexation by the US. The US and Mexican governments were at odds from the very start. One area of dispute was the actual southern border of Texas, which the US claimed was the Rio Grande and Mexico claimed was the Nueces River (Balestra, Martínez, et al. 2008: 26). War broke out over this issue in what is now Brownsville, Cameron County, in 1847. By the end of it all, the US walked away with half of Mexico’s territory. Once the border was officially drawn at what is now the end of the Lower Rio Grande Valley, the authorities followed the policies of Anglicization that were predominant in the rest of the US (Balestra, Martínez, et al. 2008: 41-42). For example, English became the language of instruction in schools in 1858 (Balestra, Martínez, et al. 2008: 42). This way, Anglos sought to cement their position of power within society (Martínez 2008: 146).

However, Spanish in the Lower Rio Grande Valley fared better than in the rest of Texas. For example, Cameron County was accommodating of bilingual education as late as the 1890s (Blanton 2004: 27). Spanish fared better in part due to the region’s proximity to Mexico, where the Spanish language is secure. The more or less constant flow of immigration North from Mexico has also contributed in this regard. Thus, in places like Brownsville, English-Spanish bilingualism has been a way of life for a long time. Over time a social structure developed where English became associated with official settings and prestige while Spanish became associated with informal settings and the absence of prestige (see González Núñez 2017: 155).

Currently, according to the 2016 American Community Survey,[3] 62% of Cameron County’s population aged five and older speak English at least “very well,” and 73% of the population speak Spanish. In turn, 1% speak a language other than Spanish or English. What this means is that the bulk of the population is English-Spanish bilingual. It is hard to gauge from the data which language is dominant among bilinguals. The survey does not indicate how well these individuals speak Spanish. Additionally, Spanish speakers are not evenly distributed, with the density of Spanish speakers decreasing as one moves northward. For example, in Rio Hondo, a town on the north of the county, 56% of the population speak Spanish at home, while in Brownsville, the county’s largest and southernmost city, 85% of the population speak Spanish at home.

Turning our attention to the United Kingdom, the area known as Gwynedd county is a local government area located on the northwest of Wales. Historically the name has referred also to a larger area covering most of northern Wales. Celtic peoples inhabited what is now Wales since before the Common Era, and their language, known today as Brittonic, would evolve into Welsh by the sixth century AD. Wales eventually came under Roman control, and after the departure of Rome, a number of Welsh kingdoms came into existence, including the kingdom or Principality of Gwynedd. Gwynedd remained independent until it was conquered by England’s Edward I in 1282-1283.

The region would be formally annexed through the Laws in Wales Acts 1535 and 1542. These laws, known as Acts of Union, were part of a process of Anglicization of the area. They banned the use of Welsh in public office and made English the only language of the courts, providing local elites with incentives to move into English (Grin and Vaillancourt 1999: 15). With the passing of the centuries, the position of Welsh continued to weaken. By the 18th century, many people in Wales “embraced English as a means of access to metropolitan and, increasingly, imperial opportunity” (Clark 1989: 226). At the same time, the Welsh language became associated with backwardness. A notorious example of this was the 1847 Reports of the commissioners of enquiry into the state of education in Wales which portrayed Wales in a very negative light and linked the supposed backwardness to the Welsh language (Kaufman 2012: 328). With the turn of the century, WWI marked the beginning of a rather rapid decline in the number of Welsh speakers, due to both the number of the dead in the war and the increased contact with the English language as regions of Wales became less isolated from England (Jenkins and Williams 2000: 3-5). This decrease in isolation, combined with accelerated industrialization, led to younger speakers of Welsh moving away from Wales in search of opportunities in England (Jenkins and Williams 2000: 5).

However, starting in the 1960s, the civil rights movement in Wales focused on the Welsh language (Jenkins and Williams 2000: 14-16). Quite gradually, from the 1960 to the 1980s, domains such as public broadcasting and public education were opened to the use of the Welsh language (Jenkins and Williams 2000: 16-17). A number of parliamentary acts indicated a shift in policy, including the Welsh Language Act 1967, the Broadcasting Acts 1980 and 1981, the Education Reform Act 1988, and the Welsh Language Act 1993. In the early 21st century, the devolution set in motion by the Government of Wales Act 1998 allowed important advancements in support of the Welsh language (Kaufmann 2012: 328). These included the passing of the Welsh Language Measure 2011 by the Welsh Assembly, which granted official status to Welsh and strengthened what were already the UK’s “most supportive State policies” for a language other than English (Dunbar 2003: 21).

Despite an overt policy regime which is probably the envy of many a minority-language communities, language planners have not been able to counter the tide of historical attrition, at least not yet.[4] In 1901, 50% of the population of Wales spoke Welsh, and in the heartland, where Gwynedd county was found, over 90% of the population spoke it (Jenkins and Williams 2000: 5). The picture is rather different in 2011. According to that year’s census,[5] 19% of the population in Wales aged three and over report they can speak Welsh.[6] (The 2013-2015 Welsh Language Use Survey calculates the number of speakers to be higher, at 24%.) The distribution of Welsh speakers is not uniform, with west Wales having the highest percentages of speakers.[7] In modern-day Gwynedd, 65% of the population report they can speak the language.[8] Gwynedd is the local authority with the highest percentage of fluent speakers of Welsh, “where almost four in every five Welsh speakers are fluent.”[9] It is worth noting that there have been no monolingual speakers of Welsh[10] since the late 20th century,[11] so these are bilingual speakers.

When one considers both Cameron and Gwynedd, it becomes evident these are two regions far removed from each other which are historically and demographically different. In terms of language, an obvious difference is that the minoritized language is differently situated on the world stage. For Cameron, the minoritized language is an international language spoken by millions of individuals just South of the border. Consequently, there is a steady influx of new speakers of the language as the US continues to be a magnet for Mexicans and Central Americans that enter the country by land. For Gwynedd, the minoritized language is in its heartland. This means, for example, that if the language struggles to be spoken in Gwynedd, it is unlikely to be reinvigorated by migration because Welsh is not a major international language.

At the same time, there are areas of commonality. Both regions were conquered by a neighboring power who imposed English as the language of administration and prestige. As a consequence, both regions saw an increase in the number of English speakers, through migration into the region and through administrative actions. Nonetheless, the minoritized language persisted in both places. Thus, in both regions the majority of speakers exhibit some level of skill in that minoritized language, and all (or most) of them are bilingual. These historical and demographic parallels make the regions comparable for the purposes of this study.

3. Translation policy in Cameron and Gwynedd

In order to address translation policy, the place where it emerges needs to be contextualized. The preceding section served that purpose, and this section will outline what translation policy looks like in the two regions in question. It should be noted that in Cameron County a two-tier level of government exists. The tier closest to local residents is the city government. The largest city in Cameron, and the county seat, is Brownsville. In addition, the county government acts as a second tier of government, taking charge of things such as keeping vital records, which city governments do not do. Across the Atlantic, Gwynedd is a unitary authority responsible for nearly all local government functions in its geographical area. Its largest town is Bangor. Because of the different local government structures, some functions which in Texas are carried out by, e.g., Brownsville (a city government), in Wales are carried out by Gwynedd (a unitary authority). The comparison between Gwynedd and Cameron/Brownsville is still valid because both represent the closest level of government to residents. Having explained this, we can now consider translation policy in these two regions.

Translation policy in the judicial and local government domains in Brownsville and Cameron County has been explored in detail in González Núñez (2017), so for purposes of this study a bird’s-eye view will suffice. An important observation is that because there are no written policy directions regarding the translation of documents or the interpreting of encounters, translation policy is developed organically through pragmatic problem-solving strategies of bilingual employees and their supervisors. In terms of local government, all documents are produced as a matter of course in English. At the county level, forms and other documents for public consumption are not translated into Spanish or into any other language. At the city level, Brownsville translates some documents for public consumption into Spanish as a matter of course, others only if a specific need is identified, and many are never translated. What little translation does take place is carried out by staff members who happen to be bilingual. They are not hired as translators but if they are bilingual, which is very often the case through sheer demographics, they can be tapped to translate. This results in ad hoc, unprofessional translation of documents. In terms of spoken interactions, neither the county nor the city government see the need for interpreted encounters, except on the rare occasions when foreign dignitaries might come on official business from neighboring Mexico. Thus, when individuals who might need to communicate in Spanish approach their local governments for services, the front line personnel who communicate with them will help them in Spanish if such personnel have the needed skills, and if not, help will be requested from some other employee. All in all, the belief becomes apparent that translation is an easily improvised tool to ensure access to some specific service.

A telling example of how much of a priority translation is for local government in this region of Southeast Texas can be found by looking at the City of Brownsville and Cameron County websites (see Images 1 and 2). Brownsville’s website, housed at, contains a menu of options on the side, a ribbon with links on the top, a search box, a number of social media icons above that, etc. Everything on the landing page is in English, and there is no obvious place to go for materials in Spanish or any other language. A thorough search of key terms using the site’s own search engine found there are documents in Spanish, but as these seem to have been produced ad hoc, they are not in a central place and range in topics from an infographic on the weather during the holidays to general safety instructions. There is a “settings” icon right next to the search box, and the user who clicks on it will find a pull-down menu which includes ten options, the ninth of which is “Translate Page.” When the user clicks on that option, a Google Translate version of the webpage loads in Spanish. This pathway to machine translation does not seem very intuitive or easy to find.

Cameron County’s website, housed at, contains a number of menus, including a ribbon at the very top. All the content in the website can be found in English, and a thorough search of key terms using the site’s own search engine revealed some materials in Spanish. These materials are not to be found in a central place either, and they too seem to be ad hoc materials ranging from instructions for court proceedings[12] to emergency preparedness materials. Unlike what happens with the Brownsville website, there is no link to machine translation anywhere. Thus, unless one knows that a specific document has been translated into Spanish and can type its name into the search box, there is no simple or systematic way to get to materials in the Spanish language.

Image 1

Brownsville’s website in June 2018. Users can have Google Translate produce a Spanish-language version by expanding the “settings” icon on the upper right-hand corner and selecting “Translate Page.”

-> Voir la liste des figures

Image 2

Cameron County’s website in June 2018.

-> Voir la liste des figures

Now we turn our attention to Gwynedd. Translation policy in Wales generally has been explored in González Núñez (2015), but the current paper will focus on Gwynedd. An important observation is that translation policy in the English-Welsh pair has developed in Gwynedd as a result of a broader policy adopted as early as the 1970s of promoting the Welsh language. In no uncertain terms, this local authority has explicitly stated its policy objective “to be an anchor for the [Welsh] language in its resurgence throughout Wales.”[13] In 1996 “Gwynedd designated Welsh as the administrative internal written and oral language” (Carlin 2014: 71). Back in the 1970s, Gwynedd had set up a Translation Unit charged with translating documents and offering simultaneous interpreting (Kaufmann 2012: 333). To help ensure the quality of the Translation Unit’s output, its translators are members of Cymdeithas Cyfieithwyr Cymru (Welsh Translators Society).

Currently, overt policy calls for a great extent of bilingualism. For example, correspondence may be in English only, Welsh only, or bilingual according to specific factors.[14] Further, all documents produced for public consumption (including posters, flyers, press releases, media campaigns, etc., whether in print or on line), are to be published bilingually or, if needed, in sister versions, one in English and another in Welsh.[15] All meetings arranged by the council are to be held in Welsh, and simultaneous interpreting will be provided as needed for English speakers who need such services to communicate.[16] Through conscious recruitment efforts, the majority of Gwynedd Council’s employees are bilingual. Thus, employees are expected to be able to produce their own bilingual e-mails, for example. However, for longer documents, the services of the Translation Unit are employed. Additionally, interpreters are used for public meetings and translated materials are produced by the Translation Unit. All in all, the belief becomes apparent that translation is an integral part of establishing and maintaining a bilingual council where the Welsh language can be an anchor for the broader region.

To what extent quality translation is a priority in Gwynedd’s delivery of services is exemplified by looking at the council’s website (see Image 3). Gwynedd’s website is housed at When users type the address, they are automatically directed to a website in Welsh. (Once cookies have stored a language preference in English or Welsh, however, the address automatically redirects to the webpage in that language.) The upper right-hand corner of the page contains a tab that reads “English.” This tab is easy to find and intuitive in its use. By clicking on it, users load Gwynedd’s English-language page. In terms of design, both landing pages are identical. Thus, what that can be done through the medium of one language in one language version of the webpage can also be done in the other language webpage. For example, one can report a problem in either language (see Image 4).

Image 3

Gwynedd’s website in June 2018. Users can load an English-language version of the website by clicking on “English” in the upper-right-hand corner.

-> Voir la liste des figures

Image 4

Both language versions of Gwynedd’s website allow residents to take any number of actions through the medium of either English or Welsh.

-> Voir la liste des figures

4. Policy determinants in Cameron and Gwynedd

When one considers the differences in translation policy between these two regions, an obvious question that arises is why the policies are so different. A superficial answer would simply be that if the two places are different, then surely their translation policies have to be different. To some extent that may be true, but there are enough similarities between these two regions that on some level local governments in both areas should have similar policies. And yet the beliefs, practice, and management of translation are so disparate, that one is drawn to consider specifically what gives translation policy such different shapes.

When asking such a question, the starting point can be a consideration of overt policy, i.e., of explicit rules that help shape translation policy. In Cameron, for example, there are legal rules that indicate to local government officials that they must provide equality of access to individuals who lack the English skills to otherwise access services.[17] These include the Civil Rights Act of 1964, as interpreted by the Supreme Court in Lau v. Nichols, and Executive Order (EO) 13166 of August 2000, Improving Access to Services for Persons with Limited English Proficiency.[18] These legal rules do not mandate translation, but they do create the right to access across language barriers. In order to eliminate said barriers, local authorities rely on ad hoc, non-professional translation provided by untrained bilingual staff.

Some legal rules in Gwynedd, in turn, are similar, but others go considerably further.[19] For example, under the Equality Act 2010, as interpreted in Mandla v. Dowell Lee, government officials must provide equality of access to individuals who lack the English skills to otherwise access services. Legally, this situates a speaker of Polish in Gwynedd in a situation similar to a speaker of Spanish in Cameron, if both of them have limited English proficiency. However, it does not do much for speakers of Welsh, because all (or, arguably, most) of them have sufficient English skills. Translation in the English-Welsh pair flows from an altogether different set of legal rules. These include the Welsh Language Measure 2011, which created a large set of rules, including service-delivery rules, to ensure that bodies such as local government councils treat Welsh and English on the principle of equality.[20] This set of rules, known officially as Welsh Language Standards, became legal duties for Gwynedd Council in 2016.[21] Some of these rules specifically mandate the use of interpreting, and they also mandate for a large number of letters, public documents, posters, signs, media campaigns, etc. to be done at least in the Welsh language. All of this necessitates a sustained translation effort which is carried out, to a great extent, through the efforts of the Translation Unit and its language professionals. These individuals provide interpreting, translation, proofreading, and guidance regarding Welsh grammar or spelling, as requested.

So the legal rules that shape translation policy in Gwynedd for speakers of Welsh are more developed than those found in Cameron for speakers of Spanish. However, that is only part of the answer, because Gwynedd often times found itself ahead of what was legally mandated. In fact, these legal rules often were catching up with policies that had already been adopted by the local government in Gwynedd. Consider this: when Gwynedd County Council[22] was established in 1974, it formally adopted a bilingual policy without being legally required to do so (Morris 2000: 577). Thus, Gwynedd found itself doing things like placing bilingual road signs long before there was any legal duty to do so (Grin and Vaillancourt 1999: 20). The councils’ language activity lacked a broader legal regime able to support it (Carlin 2014: 71), so that in many ways, Gwynedd’s policy was an act of resistance against the larger legal framework. More recently, when the Welsh Language Commissioner issued Welsh Language Standards for Gwynedd, the council already had in place most of the new requirements and in some instances even went further than what the Standards required.[23] Thus, the shape of translation policy in Gwynedd cannot be accounted for only by overt policies imposed top down from those in authority.

In order to account for the differences in translation policy as observed in Gwynedd and Cameron counties, one must consider bottom-up policy determinants. To do so, it is necessary to turn back the clock significantly. During the first half of the 20th century, faced with the decline of the Welsh language, cultural organizations such as Undeb Cenedlaethol y Cymdeithasau Cymraeg [National Union of Welsh Societies] and Urdd Gobaith Cymru [Welsh League of Youth] promoted said language at the grassroots[24] (Jenkins and Williams 2000: 10). By the 1960s the world was in a state of social turmoil, and the civil rights movement in Wales focused on the language (Jenkins and Williams 2000: 14-15). In that context, a radio lecture given by local activists Saunders Lewis galvanized a grassroots movement. All this contributed to the creation of Cymdeithas yr Iaith Gymraeg [Welsh Language Society], a non-violent, militant group which organized campaigns that managed to grudgingly get concession such as the bilingual road signs mentioned above (Jenkins and Williams 2000: 15-16). Residents of Gwynedd were not immune to all of this agitation, and local politics reflected it. In fact, by 1974 Plaid Cymru, a Welsh political party that has advocated for independence from the United Kingdom, gained control of the council (Patrick 2014: 81). The importance of these developments in eventually determining translation policy should not be understated. In fact, linguistic activity in Gwynedd “is the direct result of the deepening influence of Plaid Cymru in the organizational and linguistic practices of the local authority” (Patrick 2014: 81). All this indicates that in Gwynedd, grassroots efforts combined with the activities of local politicians where the driving force behind policy developments aimed at strengthening the position of the Welsh language. These developments helped shape current translation policy, because “[c]ontemporary translation and interpreting in Welsh have been firmly embedded in the process of attempting to regain space for the language” (Kaufman 2012: 333).

When comparing this situation with that of Cameron in the Rio Grande Valley, some differences become apparent. While the legal regime, as described above, includes some similarities, the differences in the county’s translation policy are more fully accounted for by considering the lack of bottom-up policy determinants in favor of the Spanish language. Some of this has to do with demographics. As stated earlier, most people in Cameron have some level of bilingualism, and in areas like Brownsville, monolingual speakers of Spanish outnumber monolingual speakers of English. Due to initial conquest, ongoing migration, and the geographical proximity of Cameron County to Mexico, the area has been one where Spanish and English have found themselves in robust contact for at least a century and a half (Mejías, Anderson, et al. 2002: 121). All of this contributes to no sense of attrition of Spanish speakers, at least locally. Consequently, grassroots organizations have not been galvanized around the Spanish language as a key identitarian issue.

Rather, bottom-up efforts of resistance have focused on issues such as bilingual education. A historical perspective is helpful here. In the early 20th century, the Americanization Movement sought the “formal rigid process of [immigrants] shedding foreign-ness” through US institutions such as public schools (Blanton 2004: 59). In Texas public education, the attitudes of the Americanization Movement remained prevalent up through the 1960s (Blanton 2004: 73). This resulted in things such as physically punishing children for using Spanish in school. A key figure in the push back against oppressive monolingual practices in education was Alfonso Ramírez (Dávila-Montes, González Núñez, et al., forthcoming). He was a Rio Grande Valley educator whose efforts during the 1960s included creating a local bilingual education program, for which materials in Spanish had to be developed in-house (Dávila-Montes, González Núñez, et al., forthcoming). By 1968, the federal government passed a Bilingual Education Act, which was the result of several factors, including the aforementioned activism by Mexican-Americans who felt English-only education was exclusionary (San Miguel 2004: 9-10). Thus, efforts at the grassroots and the subsequent involvement of politicians did not exert pressure on local government authorities to develop any type of translation policy beyond that which might flow from the most basic civil rights obligation to provide access.

5. Perceived problems and the shape of translation policy

As Córdoba Serrano and Dias Fouces (2018: 5-6) have stressed, public policies are, at their core, institutional responses by public authorities to perceived problems. Such an insight is helpful when considering translation management, practices, and beliefs in public settings. Specifically, in terms of translation policy one might ask: what is the perceived problem here? The shape of translation policy will depend, to a great extent, on the answer to that question. As an example of this, we will now attempt to answer that question as part of the comparative study being presented here.

As was mentioned above, Gwynedd and Cameron counties are comparable on some aspects. Both are regions that were conquered by a neighboring, English-speaking power. Both are regions where the language spoken before annexation – Welsh in Gwynedd, Spanish in Cameron – became associated with negative values, and government policies have historically pushed against it in favor of English. Both are regions with high percentages of bilingualism, to the point that over half of the population speaks the pre-English language. Both are regions where speakers of the minoritized language feel a strong emotional connection to the language as part of their identity. Even so, there are differences, including the fact that the place of Spanish in Cameron feels secure (even if subordinate to English) while there is a concern in Gwynedd that the position of Welsh is not so clearly secure.

There are important differences in the translation policies local authorities have adopted. In Cameron County, translation appears largely as an ad hoc, pragmatic solution deployed mainly by bilingual employees with no formal linguistic training in an effort to ensure access for Spanish speakers. In turn, in Gwynedd county translation appears as a sustained, planned professional activity that is integral to the county’s efforts to allow Welsh speakers to live out their lives through the medium of Welsh if they so desire.

Such different translation policies can be accounted for to some extent due to differences in top-down determinants that affect the local government. In Cameron, civil rights legislation has resulted in the obligation to find some way to provide access across linguistic barriers. In Gwynedd, there is human rights legislation to guarantee access across linguistic barriers, but this does not affect speakers of Welsh because they can communicate in English. However, there is legislation mandating that the Welsh language be treated no less favorably than the English language and there are specific legal obligations (“Standards”) to help ensure such a principle of linguistic equality is observed. These top-down policy determinants help account for what is observed in Cameron but not necessarily in Gwynedd. In the latter, bottom-up policy determinants have also played a role. Efforts at the grass roots to push back against the attrition of the Welsh language helped shape translation policy, especially as these were taken up by a regional party that came to have control at the local government level.

The different translation policies are, at their root, institutional responses to different perceived problems. In the case of Cameron, the perceived problem is the steady presence of a significant group of individuals who out of necessity must communicate in a language other than English, that is, in Spanish. This is perceived as a problem because authorities accept the narrative that English is the language of prestige and, consequently, of official functions and transactions. When someone approaches this official realm in Spanish, this poses a problem because the authorities must either turn the person away or, in the alternative, break from the prestigious use of English. Thus, the resultant policy takes a language-as-a-problem approach. The lack of proficiency in a shared language is regrettable but can be solved with relative efficiency through the use of the already available resource that is bilingual employees. (This stems from a largely instrumentalist view of language whose approach to translation can be justified through what De Schutter (2017: 18-20) describes as a privation theory of translation.)

In Gwenydd, the perceived problem is quite different. There the problem is the ongoing attrition in the number of speakers of the Welsh language, which implies the possibility at some future point that the language could disappear altogether. This is perceived as a problem because the authorities believe that local speakers of Welsh may wish, as a matter of asserting their identity, to live out their lives through the medium of Welsh. In a context where monolingual speakers of English are increasingly common, those who wish to use the Welsh language in both informal and formal contexts may struggle to find such contexts. Thus, the resultant policy takes a language-as-a-right approach. It should be noted that it is not clear there is a legal right to use the Welsh language in Wales, as the legal framework has historically been one of imposing obligations on the authorities as opposed to creating rights in the population (see Dunbar 2007). However, what matters here is the view of local authorities, which in this case is that people should have such a right, independent of the existence of such a right in legislation or case law. With this view, sustained and professional translation is deployed as one of several tools to help alleviate the lack of contexts, including official ones, in which Welsh speakers can feel secure in the use their language. (This, in turn, stems from what is a more identitarian view of language whose approach to translation can be justified through what De Schutter (2017: 20-26) describes as a theory of full translation.) As has been explained above, this approach is taken, to a significant extent, due to the pressure exerted from the bottom up.

These insights help the observer better understand why translation policy is so different in Gwyendd and Cameron. A number of factors have been considered, including history and demographics. Further, the pressure exerted by other policies, such as civil/human rights policies, has been duly accounted for. The study has also considered key bottom-up policy determinants, especially those linked to grassroots pressures on the authorities. All of this leads to the conclusion that authorities in Cameron have a minimalist, largely improvised translation policy because they approach the Spanish language as a problem while authorities in Gwynedd have an extensive, more professionalized translation policy because they approach the Welsh language as a right. In short, translation policy is shaped by a number determinants that come from the bottom up and from the top down. In this dynamics, the problem authorities are trying to solve through translation matters. And whether something is perceived as a problem depends to a great extent on the attitudes of committed individuals on the ground and their ability to exert pressure on the authorities.