Significance

Interpreters are widely used in deaf education, but the empirical record on the efficacy of interpreted education is sparse. This paper examines the use of sign language interpreters in deaf education through the lens of an emerging understanding of deaf children that separates hearing status from language deprivation (Hall 2017). We review the (lack of) evidence that interpreted education is an effective intervention for children at risk for language deprivation, and argue that there is reason to believe it may actually be harmful to children at risk for language deprivation.

Introduction

A strong link exists between education and health. A classic example is that education often leads to higher-paying jobs which in turn lead to healthier living (e.g., good health insurance, reduced stress, improved health literacy). Education also plays an important role in the development of social and psychological skills that promote better health. We present one example of a relationship between education and health whereby impoverished educational environments may actually cause or exacerbate negative health outcomes in children who are born deaf or hard of hearing (hereafter, ‘deaf’). In this paper we describe a common practice that is intended to support deaf children—the use of sign language interpreters in a mainstreamFootnote 1 classroom—and argue that improper use of interpreters can actually have serious negative health implications.

Even with available interventions and technologies, half of all elementary-aged deaf children with cochlear implants have spoken language skills below the 16th percentile (Geers et al. 2017Footnote 2). Spoken language outcomes among deaf children are highly variable and often poor (e.g., Hoffman et al. 2018). If exposed early, deaf children can develop native proficiency in a sign language like American Sign Language (ASL; Petitto 2000). Most parents of deaf children, however, are hearing (Mitchell and Karchmer 2004) and the majority of families do not use a sign language at home (Mitchell and Karchmer 2005). This means that most deaf children are at risk for language deprivation in that they have limited exposure to fully accessible languageFootnote 3 during early childhood. Language deprivation measurably harms language proficiency and/or development in domains that depend on language (e.g., cognitive, academic, socioemotional development; see Hall 2017 for a review).

Language deprivation has lasting consequences for a child’s ability to learn a first language (Mayberry and Kluender 2018), as well as academic achievement (Henner et al. 2016), socio-emotional and cognitive development (Schick et al. 2007), and brain development (Mayberry et al. 2011). There is a growing body of literature describing language deprivation among deaf children (e.g., Humphries et al. 2012; Hall 2017; Henner et al. 2018; Hall et al. 2019, Murray et al. 2019). Our goal here is to use this framework that differentiates between deafnessFootnote 4 and language deprivation in order to reconsider a common accommodation that is provided to deaf children in mainstreamed pre-k-12 classrooms: an ASL-English interpreter (hereupon “interpreter”). We focus here on the ASL-English interpreters in the United States context because the laws and practices around if and how interpreters are used in deaf education differ around the world, but parallel circumstances exist elsewhere and with interpreters of other signed and spoken languages.

Deafness and language deprivation affect children in different ways—the former is an audiological diagnosis and the latter is an acquired consequence of limited language input. The two merit different kinds of support services (Table 1). Though most deaf children are at risk for language deprivation, not all deaf children are language-deprived and vice versa. Barring additional diagnoses, deaf children have age-appropriate language proficiency in at least one language, and primarily need access to the classroom in whatever language(s) they know.Footnote 5 In contrast, language-deprived children need intervention to support language acquisition. The difference between deafness and language deprivation parallels a widely accepted distinction between speech (a means of transmitting language) and language (structured systems of symbols for encoding meaning). Deafness affects the way that language is transmitted, while language deprivation affects the entire linguistic system.

Table 1 The educational profiles of deaf vs. language-deprived students
Full size table

Deaf Education and the use of Educational Interpreters

Schools and programs for deaf children exists whose primary mission is to provide rich language environments for children to learn ASL, and are staffed by professionals with specialized expertise in educating children at risk for language deprivation. Because deafness is a relatively low-incidence population, these programs generally draw students from many neighboring (or more distant) districts. Placement in a school or program for deaf children outside of a resident district depends on a federal law that mandates that a child’s local district is responsible to provide a free and appropriate education to students with disabilities (the Individuals with Disabilities Education Act). A child’s local district may send the child to a specialized school if the neighborhood school cannot provide the necessary educational environment. However, most deaf children in the United States (80%) are educated in mainstream environments, not deaf education schools (GRI 2012). For the 39% of deaf children who use sign language in school (GRI 2012), an interpreter may be the primary accommodation.

Interpreters facilitate communication between signers and non-signers by reproducing what is said in one language in another language (e.g., English to ASL and vice versa). The practice of hiring educational sign language interpreters is unique to deaf education; spoken language interpreters (e.g., Spanish–English interpreters) are rarely provided in k-12 schools. In many mainstream classrooms, there is one deaf student in a room of hearing non-signers. Interpreters in educational environments often work alone, or occasionally in a team of two. The interpreter(s) and deaf student(s) and sometimes an itinerant deaf educator are often the only people in the school who know ASL. The same interpreter may be hired to interpret all school-related activities for a year, and it is not uncommon for an interpreter to follow a student year after year. K-12 interpreting accounts for 21% of the interpreting work nationwideFootnote 6 (RID 2016).

The move to use interpreters as a primary accommodation for deaf students is consistent with a trend internationally across contexts to “favor giving access to services through sign language interpreters instead of via language-concordant services, where the client and service provider use the same language” (see De Meulder and Haualand 2019 for a review). The trend of “including” deaf students via interpreters is puzzling, as mediated access to education clearly contradicts the spirit of inclusion intended by most legislation (e.g., the Individuals with Disabilities Education Act, the UN Convention of the Rights of Persons with Disabilities; Murray et al. 2018a).

Mismatch Between the Needs of Language-Deprived Students and Interpreting Services

While interpreters may or may not be a reasonable accommodation for deafness, we argue that there is no evidence that interpreters can singlehandedly meet the unique needs of children who are language-deprived. Interpreters are generally not trained in language education, and are not afforded the time or authority to provide such instruction. Though infants are able to learn language passively, there are limits to implicit language learning among older children (Ellis 2009; Hulstijn 2005; Schmidt 1992). Older children learning a second language benefit from explicit language instruction (Spada and Tomita 2010). We know of no empirical evidence that children can reliably overcome the consequences of language deprivation without explicit, immersive language instruction.

An interpreter may be the sole source of sign language exposure for a child. Children learn language best from multiple language users and particularly benefit from immersion in a peer group of language users (Gagne 2017; Swain et al. 2002). Interpreted interactions cannot substitute for direct communication with peers (Winston 1994). Peer relationships are especially important for children who have language deprivation, as language deprivation has negative effects on children’s social development. Further, as outlined below, interpreters have varying, often low, levels of sign language proficiency. We know of no empirical evidence that children can reliably overcome the consequences of language deprivation via a single language model, even if the interpreter is a highly proficient signer but particularly if that person is not a proficient signer.

Interpreted language is sometimes the sole source of language exposure for a child. As we will describe in more depth in the following section, interpreted content has a high error rate (conservatively 26–58% of utterances; Nicodemus and Emmorey 2013). This is problematic because it requires students to mentally fill in the gaps to compensate for miscommunications. Even an error-free interpretation is not the same as direct communication; interpretations often include long pauses or other disfluencies. We know of no empirical evidence that children can reliably overcome the consequences of language deprivation primarily from interpreted language.

Schools have a legal responsibility to provide a free and appropriate education, and Endrew v. Douglas County specify that IEPs must be “reasonably calculated to enable the child to make progress appropriate in light of the child’s circumstances.” Though some might argue (and we would agree) that using interpreters is better than nothing, Endrew v. Douglas County makes clear that the legal standard is greater than de minimis – rather, students must be able to make reasonable progress. We know of no empirical evidence that children with language deprivation can make reasonable progress in language acquisition primarily through interpreted education.

Using interpreters in lieu of educational supports tailored to language deprivation may exacerbate the symptoms of language deprivation. There is a linear relationship between age of language acquisition and language proficiency: the longer a child spends without language, the poorer their outcomes are likely to be (e.g., Pénicaud et al. 2013). If a child is unable to learn a useable first language via an interpreter, years spent relying solely on an interpreter to support language acquisition may prolong the period of language deprivation, thereby worsening outcomes. In this case, progress toward learning goals may not even be de minimis. Instead, the illusion that the child is being provided access may prevent educational teams from doing more to support language acquisition, further extending the period of time the child spends in an impoverished language environment.

As such, we argue interpreters are not an appropriate intervention for language deprivation. To be clear, we are arguing to limit the use of ASL interpreters – not the use of ASL. As has been described elsewhere (Napoli et al. 2015), there are many good reasons for children to use a sign language not least because spoken language outcomes are unpredictable and often poor (e.g., Geers et al. 2017; Hoffman et al. 2018). We believe that interpreting can be an effective means of providing access for deaf children, even for children with language deprivation, when used in combination with other services (e.g., to provide access to a course or sports team). Our contention is simply that interpreters should not replace interventions for language deprivation. In the following section, we review some considerations in choosing to use interpreters, and conclude with suggestions for alternatives to educational interpreters.

Considerations When Considering the Hiring of an Interpreter

Interpreters vary both in ASL fluency and interpreting ability. Thirty-one states require that interpreters score at least a 3.5 on a 0–5 scale on the Educational Interpreter Performance Assessment (EIPA). The EIPA characterizes interpreters with such scores as having “knowledge of basic vocabulary, but lack[ing] vocabulary for more technical, complex, or academic topics,” and “sign[ing] in a fairly fluent manner” with some errors in sign production, grammatical production, and notes that these interpreters may require repetition or assistance in order to understand ASL. Eleven additional states have slightly stricter requirements (4.0 out of 5), but interpreters may still “have difficulty with complex topics or rapid turn taking” (EIPA Rating System). In practice, the competence of interpreters may be even lower. In a study of the 2,091 interpreters across the US who took the EIPA assessment, more than 60% had scores lower than a 3.5 (Schick et al. 2005). In another early study, some 65% of educational interpreters surveyed were not certified, and 61% were not proficient or only somewhat proficient in sign language at the time of hire (Jones et al. 1997). Another study of two rural states found that of 63 interpreters, none were certified (Yarger 2001). Many states have a “provisional” status that allows people with even lower scores to work as interpreters temporarily (e.g., Montana allows those with a competency level of 2.5 to work for 3 years; Johnson et al. 2018). The EIPA itself recommends that people with such low skills should not work in classroom settings, as they have “only basic sign vocabulary” and their “lack of fluency… often interfere[s] with communication” (EIPA Rating System).

In addition to varying levels of sign language proficiency and cultural competence, the process of interpreting introduces additional opportunities for miscommunication. Nicodemus and Emmorey (2013) found that expert interpreters (national certified with more than ten years of experience) had a 25.8% error rate, and novice interpreters had a 58.3% error rate in interpreting three-minute personal narratives and informational texts. This error rate may be even higher in more linguistically demanding scenarios (e.g., trigonometry). It may be especially difficult for students to mentally fill in the gaps of an error-filled interpretation because the topics covered in school are, by design, subjects they are not deeply familiar with.

Lastly, accuracy rates steeply decline due to physical and mental fatigue after about thirty minutes of interpreting (Brasel 1976). Interpreters who work in other contexts (e.g., higher education, professional settings) generally work in pairs and trade off every twenty minutes to preserve the quality of interpretation. In contrast, it is not uncommon for educational interpreters to work continuously all day alone. The error rate in educational interpreting may be even higher than 25.8%-58.3%, as error rates are based on about a three minute interpreting sample.

We add one final note on a common (mis)use of interpreters in educational settings focusing on language assessment. The first step in addressing language deprivation is identification, which, at minimum, entails testing children’s varied language proficiency (e.g. in both home and community languages). Without standardized assessments of language deprivation, schools may be left to use subjective impressions of the child’s language skills. An interpreter may be the only member of the educational team that knows ASL and could comment on the child’s ASL proficiency. Interpreters, however, generally have limited ASL proficiency, are not trained to provide ASL assessment, and are likely to be ill-prepared to make an informal evaluation (Schick et al. 1999, 2005). Even if interpreters are aware that children have limited language proficiency, they may not have the training or authority to inform school personnel. The code of ethics governing sign language interpreters in the United States may further discourage interpreters from communicating concerns with school personnel, as it indicates interpreters should not “provide counsel, advice, or personal opinions.” Practically, interpreters are also unable to share their thoughts in a meeting if they are simultaneously interpreting the meeting.

Because of these limitations, we believe that interpreters are a non-optimal educational option for children who are language-deprived. While interpreters may be a more appropriate accommodation for deaf students who are not language-deprived than those with language deprivation, interpreted education may not be equitable educational environment even for children who are not language-deprived. See Murray et al. (2018b) and Thoutenhoofd (2005) for arguments here.

Conclusions for Practice

With an emerging understanding that deafness and language deprivation are separate but related conditions, it is incumbent upon educational teams to 1) evaluate deaf children for language deprivation, 2) identify the needs that arise as a result of language deprivation, and 3) provide services or environments that meet those needs. While every child’s needs are unique, children with language deprivation generally need educational environments that are rich with accessible linguistic input and meaningful linguistic exchanges with both adults and peers, convivial and accessible social environments, and staffed by educators with expertise in language deprivation. Until robust diagnostic tools for language deprivation are available, we urge schools and families to act proactively and assume deaf children need language education including explicit instruction from professionals who have training in language education, and immersive language environments where children have ample opportunities to interact with ASL fluent peers and adults..

The most straightforward alternative to educational interpreters is to take advantage of the many schools and programs nationwide that are designed to provide a rich, immersive language-learning environment for language-deprived children. Bilingual schools for deaf children are staffed by people who have expertise, often advanced degrees, in educating language-deprived students. By congregating deaf students, these schools can provide students a social system replete with peers and adults who can communicate directly in ASL. These environments play a crucial role in language and social development, and can affirm students’ socio-cultural identity.

Children’s local districts may be reticent to send their students to specialized schools for a number of reasons (e.g., expense, the misguided notion that inclusion entails sending children to neighborhood schools; Trahan et al. 2018), preferring to educate their students in-house. It may be possible to hire staff with expertise in supporting delayed first language acquisition, but we expect it will be challenging for most districts to provide the necessary social and linguistic environment, as doing so would require a critical mass of fluent signing children in the local district. As such, many districts will have trouble independently providing the necessary environments.

There may also be geographical and/or financial constraints, and require more creative solutions. One promising avenue is the use of remote learning where students from around the country could participate in specialized programming online. Video conferencing may enable students to access high quality instruction and a classroom of fluent signing peers that would otherwise be unavailable in their local district. However, more research is needed to examine the effectiveness of distance or computerized education for providing language access for deaf children. Additionally, districts could build capacity locally, hiring fluent deaf signers to participate as language models in mainstream classroom activities (e.g., as teachers assistants), and/or provide teacher training to people who are fluent in ASL.

We are calling for systemic change in practice—individuals cannot singlehandedly make the necessary changes. Schools should consider using educational interpreters only in exceptional circumstances. Interpreters should not accept work in educational contexts without clear evidence that doing so would not hamper students’ ability to thrive. Interpreters should counsel those requesting their services about safer alternatives, and should work to counter the misconception that educational interpreting is an acceptable accommodation for most deaf children. Professional organizations for sign language interpreters should openly condemn the practice of educational interpreting except in rare circumstances. Professional organizations could lobby for legislation to minimize use of educational interpreters. Certifying bodies could cease to offer credentials for educational interpreters, which lend credence to this harmful practice.

With interpreters, deaf people can more easily engage in a host of activities that would otherwise been inaccessible, which ultimately improves health and wellbeing across the lifespan. At the same time, interpreters can give administrators and parents an “illusion of inclusion” (Winston 1994; Russell and McLeod 2009) leaving deaf children without the support they need to mitigate the consequences of language deprivation, and in some cases putting them at further risk by prolonging the period of time spent without a complete first language. These children need specialized interventions tailored to language deprivation, not just interpreters. As such, interpreters should not be used as an intervention for language deprivation.