Informed consent in research on second language acquisition

Abstract

The practice of securing informed consent from research participants has a relatively low profile in second language (L2) acquisition research, despite its prominence in the biomedical and social sciences. This review article analyses the role that informed consent now typically plays in L2 research; discusses an example of an L2 study where complex issues of informed consent surfaced; and summarizes debates about informed consent that are underway in other disciplines, but which so far have been little recognized in scholarship on L2 acquisition.

Keywords

informed consent Institutional Review Board research ethics in second language acquisition

I Introduction

Informed consent as a cornerstone in the ethics of scientific research is almost 60 years old. However, it has not yet fully come of age in research on second language acquisition. This review article calls attention to evidence that the full complexity of informed consent has not been adequately appreciated or integrated into research on how people learn second languages (L2s).¹ We suggest that L2 research may be improved by increasing awareness of several robust on-going debates about the nature and role of informed consent.

We begin by introducing the concept of informed consent in its historical context. Next, we survey the role that informed consent now plays in L2 research, gathering data from textbooks on research methods; the websites of professional organizations; and a sample of recently published empirical studies of L2 acquisition. We conclude with an exposition of how scholars in other social-scientific fields have developed, sometimes challenged, certain facets of informed consent in ways that are worth the attention of L2 researchers. Integral to our discussion is a case study of L2 research, included as Appendix 1, that problematizes conventional practices of informed consent and illustrates why the notion needs to be better understood and better adapted to cross-cultural contexts.

Our perspective derives from the academic culture that both of us know best, that of the twenty-first century United States. A full international analysis of informed consent exceeds the scope of this review article. We occasionally refer to informed consent in the intellectual life and research practices of other countries, however, because part of the difficulty – and the intrigue – of the concept derives from the questions it raises about what constitute transcultural ethical standards.

II Emergence of informed consent as an ethical principle in research

Informed consent is foundational to research that involves human beings. Arguably, it is the heart and soul of the ethical dimension of science. In many accounts, revelation in 1948 of the horrors of Nazi pseudo-scientific experimentation led jurists presiding at the Nuremberg trials to articulate principles which, contrary to Nazi practices, defined legitimate medical experimentation. Those principles require that any participant in an experiment offer voluntary, legally competent, informed, and comprehending agreement. In 1957, these criteria were elaborated, inscribed into the U.S. legal code, and labeled with the expression ‘informed consent’, in the course of settling a medical lawsuit in California. That lawsuit clarified that a patient’s consent to treatment can only be valid when it is fully informed; that is, after a doctor has explicitly disclosed all pertinent information about the ‘nature, consequences, harms, benefits, risks, and alternatives’ of the proposed treatment (Faden and Beauchamp, 1986: 126). A further milestone in the history of informed consent was the Helsinki Declaration of 1964, which attempted to define international, although not legally binding, standards for ethical medical research.

These events feature prominently in most historical sketches of informed consent. Faden and Beauchamp’s classic 1986 monograph A history and theory of informed consent provides more complete discussion. The text begins by addressing the conceptual basis of informed consent in western moral philosophy and legal thought. Part II provides a history of the notion as it developed in medicine, law, scientific research, and U.S. federal policy. Part III works out a theory of informed consent. First, Faden and Beauchamp define the concept of ‘autonomous action’ as resting on the conditions of intentionality, understanding, and ‘noncontrol’ (that is, independence of control by others; p. 256). They then define informed consent as a patient’s (or research participant’s) autonomous action to authorize a particular procedure or intervention.² The book concludes with a chapter that closely analyses what is entailed in meeting the criterion of understanding, and another chapter on the criterion of noncontrol.

Like most literature on informed consent, this account draws its examples largely from biomedical research, where informed consent aims to set a clear and high standard for patients’, or research participants’, understanding of the risk of material physical harm. Faden and Beauchamp also narrate how informed consent was adopted from medicine into the social sciences, taking the field of psychology as a microcosm. Here, the stakes are typically – but not necessarily – lower, and potential harm to research participants is less likely to be physical than emotional or social: boredom, confusion, tension, embarrassment. The American Psychological Association began formulating a professional code in 1938. By the early 1950s psychologists had articulated a concept of informed consent, without using that term (Faden and Beauchamp, 1986: 167–71). During the 1960s, there was vigorous public debate of the ethics of deceiving participants in psychological studies, or of providing them with incomplete information in cases where full disclosure would undermine the validity of an experiment. By 1973, the American Psychological Association had published a statement of research ethics requiring researchers to treat participants with openness, respect, and honesty. It prominently addresses, if not resolves, the question of deception.

Because deception threatens the spirit of informed consent, the continuing controversy surrounding deceptive research indicates the extent to which social scientists take informed consent for granted. It became a matter of law in the U.S. with the development of Institutional Review Boards (IRBs).³ From 1974, the federal government, acting through the National Science Foundation, required all U.S. institutions that carry out federally supported research to establish IRBs, which review and approve of any study involving human participants before data are gathered. Among their charges is ensuring that every participant documents his or her informed consent by reading and signing what is often a long, fine-print, document written in quasi-legal language. As a minimum, informed consent forms must explain how participants’ privacy is protected; indicate that participants may withdraw at any point without prejudice; provide researchers’ contact information; depict potential benefits to participants; and, crucially, describe any conceivable risks of harm involved in the research (at least in the conception of the culture that generates these forms). Once signed, the forms materialize participants’ willingness to be studied.

In addition to ensuring that the standard of informed consent is met, IRBs weigh the costs to participants against the potential benefits to scholarship in cases where deception or partial disclosure is involved, then advise researchers accordingly. IRBs are also responsible for judging which research projects can be exempted from the time-consuming and meticulous process of full IRB scrutiny, on the grounds that they do not pose a substantial risk of harm. Of particular salience to the study of L2 acquisition is the fact that IRBs usually identify educational research as exempt from full review. Likewise, research based on the ‘collection or study of existing data if publicly available’ (http://www.nsf.gov/bfa/dias/policy/hsfaqs.jsp#exempt), such as corpus data, is exempt from full review (even, in this case, exempt from the burden of establishing informed consent). Researchers, however, cannot assume that their work meets the criteria for exemption: a local IRB grants that status, case by case. Ethnography is another method that receives special treatment. Ethnographic research must be approved by an IRB, and may or may not be deemed exempt. In either case, procedures for establishing and recording informed consent in ethnographic studies may be tailored to the relevant cultural context (http://www.nsf.gov/bfa/dias/policy/hsfaqs.jsp#ethno). An additional responsibility of IRBs is to ensure that extra precautions are met when participants comprise members of certain ‘vulnerable’ classes of people for whom higher criteria for informed consent hold: children, prisoners, pregnant women, persons with physical or mental disabilities.

This sketch depicts the policies set by the National Science Foundation, to which research carried out in the United States or conducted by researchers affiliated with U.S. institutions must conform. Local IRBs are responsible to ensure that conformance. However, different institutions’ IRBs vary widely in how they apply these policies, for example, in how they define ‘vulnerable’ classes of people, or whether they accept alternative means of establishing informed consent. Moreover, other countries have developed their own standards for ethical research, which often include a concept of informed consent. Clough et al. (2013) review research practices in Australia, Azerbaijan, Iran, the Philippines, and South Africa. They conclude that protection for participants is sometimes inadequate and poorly adapted to local circumstances, a situation that obtains equally in the U.S.

III Informed consent in L2 research

The notion of informed consent emerged first in the biomedical sciences, and retains ties to its origins. Perry (2011), for example, analysed 32 U.S. universities’ IRB websites and found that ‘social science IRBs still use language reflective of medical models of research’ (p. 899), with little adaptation to the study of humans as social-psychological beings. This gap is more conspicuous in some L2 research contexts than others; for instance, dealing with what Van den Hoonaard (2001: 24) calls ‘the hard architecture around current research-ethics review’ may be especially problematic for ethnographic research, or where a researcher partners with L2 learners who have had interrupted formal schooling (Bigelow and Pettitt, 2015). Securing informed consent may seem like overreach for the majority of L2 studies that are deemed exempt from full IRB review, since exemption acknowledges that participants face no substantial threat. Nevertheless, the status quo (as defined by IRBs, supporting institutions, professional associations, and funding sources) insists that informed consent be met as a guarantee that participants understand, in advance, the nature of the research and any potential harm entailed in their participation.⁴

To what extent, then, has research in L2 acquisition that involves human participants incorporated informed consent into its standard procedures? We analyse three sources of evidence bearing on this question: (1) textbooks on research methods, used to transmit discipline-specific practices to novice researchers; (2) the websites of professional organizations, which beginners as well as established scholars might consult to clarify professional standards; (3) a sample of recently published empirical studies of L2 acquisition, as a measure of the extent to which published literature explicitly incorporates informed consent.

1 Textbooks

We searched 11 textbooks and two edited collections of essays on research methods for their exposition of informed consent. All were addressed to beginning researchers in L2 acquisition, and published in English from 1980 onwards.

With the exception of Seliger and Shohamy (1989), who devote nine pages to summarizing the central concerns of IRBs (participants’ freedom from coercion; masking their identity; maintaining confidentiality of the data; debriefing; a principal investigator’s responsibility for co-authors), informed consent is absent from L2 research methods textbooks until the mid 2000s (e.g. Brown and Rodgers 2002; Hatch and Farhady, 1982; Hatch and Lazaraton, 1991; Johnson, 1992; Nunan, 1992). In general, training in L2 research methods as reflected in these texts seems devoid of reflection on ethical issues, and little attuned to perceiving complexity in relationships between researchers and research participants. For example, Hatch and Lazaraton’s (1991) discussion of ‘Collecting research evidence’ (pp. 28–32) seems exclusively to assume a researcher’s perspective, as when they remark that ‘If the method you use is dull or frightening or boring or takes too long, it is unlikely that your subjects (Ss) will be motivated to perform as well as they might’ (p. 29). Hatch and Lazaraton describe an experiment that exposes learners to electrical currents, or flashes of light that cause eye blinks; they concede that the latter method might be ‘not so frightening’ compared to the first (p. 30), but their general orientation invests participants with little agency and is not aligned to participants’ experiences. Instead, they prioritize pursuit of researchers’ goals and objectify L2 learners, both conceptually and syntactically. For example, a characteristic sentence reminds researchers-in-training to ‘consider exactly what information must be obtained from the respondent’ (p. 31). Absence of discussion of informed consent is therefore not surprising.

With the publication of Mackey and Gass (2005), however, informed consent entered into training in L2 research methods. It is the first substantive topic addressed at the beginning of Mackey and Gass’s Chapter 2 (pp. 25–43). This section ranges over the rationale, function, and operation of IRBs; methods of obtaining informed consent, including from children or non-English speakers; deception and incomplete disclosure; how to prepare a research protocol for IRB review; what a consent form looks like. Following publication of Mackey and Gass (2005), textbooks on L2 research methods (e.g. Blom and Unsworth, 2010; Dörnyei, 2007; Phakiti, 2014) as well as collections of essays with a similar pedagogical goal (e.g. Mackey and Gass, 2012; Paltridge and Phakiti, 2010) routinely devote substantial attention to ethical issues, inevitably highlighting the importance of informed consent.

Graeme Porte’s Assessing research in second language learning, written as a guide to evaluating L2 research, pinpoints the sea change in consciousness of ethical issues that took place in the first decade of the 2000s. Porte’s 2002 first edition makes only scant reference to research ethics in a few parenthetical comments and footnotes; informed consent does not appear. In contrast, the 2010 second edition dedicates a separate section to several facets of research ethics, including informed consent (pp. 99–104). Moreover, on page xxv Porte (2010) notes that ‘The increasingly important critique of research ethics in L2 acquisition data gathering and research’ was one motivation for revising his text. After 2005, it is fair to say that recognition of the importance of informed consent had arrived in L2 research textbooks.

2 Websites of professional associations

While informed consent infiltrated L2 research methods textbooks in the middle of the first decade of the 2000s, it does not yet have much presence in another typical repository of disciplinary standards: the websites of professional associations. As Coady and Bloch (1996) point out, articulating a shared set of ethical guidelines is one way a discipline builds its own self-consciousness. It is striking that sibling fields like language testing and TESOL have developed their own ethical standards (Thomas, 2009), but no organization that identifies itself specifically with L2 research has done so. Moving to what is arguably the superordinate discipline of applied linguistics, one sees that neither the Association Internationale de Linguistique Appliquée, nor its national affiliates in the US, Canada, Ireland, or New Zealand have posted ethical guidelines on their websites. The British Association for Applied Linguistics (BAAL) is an exception: the BAAL website provides ‘Recommendations for good practice’, including informed consent, written in 1994 then revised in 2006. That same document, with revisions, is posted on the website of the Applied Linguistics Association of Australia.

L2 researchers needing information about informed consent might therefore have to seek discussion of its complexities outside the field. At best, they might locate the BAAL standards, or, at one step removed, those of the American Educational Research Association, or the guidelines posted on the websites of national organizations of psychologists (e.g. in the US, Britain, Australia, Canada). There are, of course, resources for researchers available from trans-disciplinary institutions such as the European Commission’s Research Directorate, Canada’s Tri-Council Panel on Ethics, or (for the U.S.) the National Science Foundation. But the general orientation and presuppositions of these institutions are even more attenuated from the context of L2 research. If informed consent were a straightforward procedure, untroubled by discipline-specific or socio-cultural complexity, that fact might not be of much consequence. However, as the narrative presented in Appendix 1 to this article illustrates, informed consent entails quite demanding and culturally sophisticated communicative tasks, which can best be achieved if the procedure is tailored as closely as possible to a particular research objective and a particular participant population. For that reason, the absence of models for, or even much discussion about, informed consent within L2 research is a missed opportunity. For professional associations to articulate discipline-specific standards would not settle all issues. But the conversations that might ensue would raise awareness of those issues and their intractability.⁵

3 Published journal articles

If L2 research methods textbooks have only recently begun to address informed consent, while professional organizations lag behind, published literature in L2 scholarship rarely acknowledges informed consent. We analysed four years of empirical studies of L2 acquisition published since 2011 in three well-established journals: Language Learning, Vol. 61.3 (September 2011) through Vol. 65.2 (June 2015); Second Language Research, Vol. 27.3 (July 2011) through Vol. 31.2 (April 2015); Studies in Second Language Acquisition, Vol. 33.3 (September 2011) through 37.2 (June 2015). Removing from consideration research based on L2 corpora, re-analyses of earlier research, and work that does not introduce novel language data, we examined the resulting corpus of 259 empirical studies conducted in 35 countries. We searched each of these studies for reports that their authors followed procedures of informed consent prior to gathering data. Of those 259 studies, only 42 (16.2%) specified that informed consent was obtained from the participants. In contrast, 193 (74.5%) make no mention at all of informed consent, or of engagement in ethical review by an IRB-like entity. This includes studies funded by major sponsors whom one might expect to require strict adherence to scientific ethics (the US National Science Foundation; Canada’s Social Science Research Council; the Max Planck Institute; the Australian Research Council; the French National Research Agency). It also includes studies carried out in countries with strong traditions of IRB-like oversight of research. To mention only countries where 10 or more studies were executed, the majority of L2 research in our sample does not register any provision of informed consent to participants: for the U.S., 74 out of 100 studies (74%) made no reference to informed consent; for the UK, 21/25 (84%); Canada, 19/23 (82.6%); the Netherlands, 12/20 (60%); Japan, 8/15 (53.3%); China, 11/12 (91.7%); Spain, 9/11 (81.8%). This includes numerous studies involving minors – elementary school-aged children, children as young as three years old, and even infants – where there is no indication that informed consent was obtained.

4 On the absence of reference to informed consent

It is possible that some authors of the studies we analysed passed their research plans through full ethical review and secured informed consent from their participants, while neglecting to report those facts in print. Anecdotal evidence gleaned from discussion with colleagues suggests that this is fairly common, at least among the U.S.-based scholars. It is also possible that some authors included informal procedures of securing participants’ consent in their research design, without meeting the exact professional standards defined by IRBs. It is relevant that 24 of the 259 studies we examined (9.3%) fell into an ‘other’ category, in which authors indicated that they established some kind of advance agreement from participants, but in our judgment that agreement would not meet Faden and Beauchamp’s (1986) standards for being both ‘informed’ and ‘consensual’. This included cases where, for example, researchers negotiated access to school-aged participants by describing the study to the school principal, but did not consult with, or inform, the participants themselves; or where children’s participation in a study was appropriately solicited from parents, but there is no indication of what, if anything, parents – or their children – were told about the content of the study, or about their option to decline or to withdraw.

Therefore some of the studies in our sample may, in fact, have passed through ethical review, without that fact being reported. Other studies may have carried out some kind of informed consent outside of formal review. Still others may not have attended at all to participants’ ‘intentionality, understanding, and noncontrol’ (Faden and Beauchamp 1986: 256). In any of these cases, to omit reference to ethical review leaves a reader uncertain about the basis of the resulting data. It also signals the extent to which informed consent is not fully integrated into the consciousness of authors, of reviewers of submissions to journals, or of journal editors. A lot stands behind the simple declaration ‘This research was reviewed and approved by the IRB of University X’. If informed consent matters in L2 research – as it does in other social sciences – the absence of that declaration is worth noting.

Fortunately, our data suggest a trend toward greater recognition of informed consent. Combining the results across all three journals, blocked into chronologically ordered four-issue quartiles, the proportion of studies that report obtaining informed consent increased from 2011 to 2015: 12% (for mid 2011–mid 2012); 15.3% (2012–13); 17.5% (2013–14); 21% (2014–15). We fit a regression model with percentage of explicit reference to informed consent as response and quartile as the predictor. A borderline significant positive relationship emerged (β = 0.008333; p = 0.0671), suggesting that although these results do not reach the level of statistical significance, reference to informed consent increased from mid-2011 to mid-2015.

Our conclusion is that despite the high profile of informed consent in the practices of other social sciences, and despite an emerging recognition of its role in textbooks on L2 research methods (compromised by its only faint presence on the websites of professional associations), informed consent is not at present a fixture in contemporary research on L2 acquisition. We leave open the question of to what extent researchers secure informed consent without reporting that fact. The scarce presence of informed consent may change as the generation trained in L2 research methods after 2005 assumes leadership in the field, if the trend in the direction of securing and reporting informed consent continues. If so, what new issues might that trend raise?

IV Informed consent: ‘It’s complicated’

Appendix 1 presents a narrative illustrating some of the ethical challenges posed by research on L2 learning. The narrative depicts Pettitt’s negotiation of informed consent in an ethnographic study. The participants’ status as refugees with diverse kinds of multilingualism, and with little experience in western educational settings, throw into stark relief the complexity of informed consent.

In our discussion below, we introduce several on-going debates within the social sciences that bear on Pettitt’s experiences. These debates raise issues relevant to L2 research, and suggest resources it might tap, if we were to take informed consent more seriously, and to incorporate it more fully into our research practices.

1 Is informed consent truly informed?

The conventional means of documenting informed consent is to ask participants to sign and date a consent form, printed on institutional letterhead, following oral explanation of its contents. However, a number of scholars have questioned whether this proves that participants’ consent to participate is truly informed. McNutt et al. (2008) discuss two studies, involving more than 300 survey participants (all speakers of ‘English or Spanish’ [p. 90], mostly high school graduates), who were observed reading 665- or 1,462-word informed consent forms. Seventy-one percent of those confronted with the shorter form appeared to read it and be ready to sign in 30 seconds or less; 55% of those with the longer form, in one minute or less, intervals wholly insufficient to process the forms’ contents.⁶ Agre et al. (2003) review eight studies that explored how informed consent might be made more engaging and meaningful to participants in clinical trials (e.g. use of video; enhanced print; decision aids). None proved very effective. The authors conclude that ‘many individuals made their decision to take part in a trial before the consent process occurred’ (p. s18), and therefore were impervious to attempts to increase their engagement in the procedure. Breese et al. (2007) document that participants with less formal schooling had significantly lower levels of comprehension of informed consent forms. Participants whose native language was Spanish or Vietnamese (rather than English, the dominant environmental language) also showed lower comprehension, even when the forms were translated into their native language and administered by bilingual staff. Kripalani et al. (2008) explore whether a teach-back exercise can assess the extent to which informed consent actually succeeds at informing participants.

These studies indicate that some researchers lack confidence that conventional informed consent provides research participants with the information they need, in a form that makes sense to them. There is also evidence of a converse problem: Walkup and Bock (2009) inquired into what participants want to know about a study in advance, which turns out to be much less than provided through conventional informed consent. Although Walkup and Bock do not advocate reducing the information supplied to participants, they cite research showing that elaborate assurances of confidentiality and harmlessness (especially couched in the legalistic language of informed consent forms) can actually backfire, decreasing participants’ confidence and comfort.

In any kind of research, it is vital that researchers fully inform participants about how involvement in a study may affect them. Although the risk of harm in L2 acquisition studies is typically very low, the kinds of metalinguistic tasks that participants may perform – providing grammaticality judgments, manipulating novel words, mapping sentences onto pictures, pushing a button when they recognize a specific sound, allowing their eye movements to be tracked – may nonetheless be perceived as intrusive or challenging or may threaten participants’ sense of language competence. So may other methods of studying L2 acquisition, which may involve participants being interviewed, having their spontaneous speech audio- or video-taped then analysed, or having texts they have created scrutinized. Participants may be concerned about being singled out in a manner that contravenes their solidarity within a group, or about public exposure of their performance relative to that of other participants. They may also perceive sources of potential harm that do not occur to the researchers. Therefore improving informed consent so it provides just the right amounts and kinds of information is important to L2 researchers as to other scholars, so that L2 learners taking part in research feel confident that they are voluntarily assuming no more than they are willing to assume. Moreover, in the course of a single study, participants may need additional information as they go along, adding complexity to what it means for informed consent to be genuinely informative.

2 Is informed consent truly consensual?

To the evidence that standard procedures of informed consent do not necessarily result in truly informing participants, Corrigan (2003) raises a correlative issue, namely, whether those procedures provide participants with an authentic choice about their involvement. In a series of interviews with participants in clinical trials, Corrigan found that they showed very high levels of trust in the institutions and individuals who organized the trials, despite low levels of understanding of the nature of the trials themselves and of their role as participants. Many misunderstood that an invitation to join a clinical trial indicated that doing so was part of their treatment, which Marshall (2006: 26) calls the ‘therapeutic misconception’. Some believed, despite multiple reassurances to the contrary, that refusal to participate would disrupt the provision of medical services on which they depended. (In L2 research, an analog is the potential dependence of student-participants on teacher-researchers.) For these reasons, Corrigan questions whether informed consent succeeds in communicating to participants that they may make a genuine, voluntary, choice. Reinharz (1993: 78) offers a more radical critique: that securing informed consent in the classic, formal, written manner is necessarily coercive insofar as it implicitly aligns researchers with institutional powers that overwhelm all but those most robust individuals who are willing and able to resist authority. Just as L2 researchers need to be attentive to whether informed consent is truly informed, they also need to attend to whether it is truly consensual; that is, whether the procedure allows participants to freely express their desire to be involved or not.

3 Who can grant informed consent?

A third, and very active, debate in the social sciences regarding informed consent has high relevance to cross-cultural L2 research. The key issue is recognition that informed consent rests on pervasive western cultural notions of the autonomy of individuals and contractual nature of social relations that adults choose to enter into. Levine (1991), among others, discusses challenges to the universality of these notions. Robinson (2010) explores what informed consent means for ‘analog people in a digital world’ citing linguistic fieldwork in the Philippines; Perry (2011) discusses its relevance to her research with Sudanese refugees in the United Sates. Problems of fitting the square peg of informed consent into the diverse round holes of different cultures’ conceptions of individuality and social relationships also emerge in research in Japan (Levine, 1991: 210) and China (Corrigan, 2003: 770), to cite two countries where learners are frequently involved in L2 research. A misfit can show up in various ways: cultural norms may expect females to consult male relatives before providing informed consent, subverting the notion that each person makes an autonomous choice (Marshall 2006); a community may reject research that demands informed consent from individuals because it lacks a provision for recognizing ‘the moral status of collectives’ (Kaufman and Ramarao, 2005: 164; see also Clough et al., 2013); the officious, impersonal, language of informed consent may be perceived as ‘alienating and dehumanizing’ (Gostin, 1995) to the extent of reducing potential participants’ sense of their own agency, the very attribute the procedure was intended to protect.

4 How is informed consent materialized?

In the expectations of many IRBs, a signed and dated paper is the gold standard for materializing informed consent. However, when IRBs (or their international analogs) demand that form of documentation, they may overshoot the actual standards they are directed to uphold. For example, the National Science Foundation (NSF) guidelines state that oral consent may substitute for signed forms, for example with participants at beginning levels of reading (http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.html#46.117(c)). Documentation on video is also an option (Benitez et al., 2002). Nevertheless, Haggerty (2004: 411) adverts to how in an attempt to maintain bureaucratic consistency, local IRBs may become ‘divorced from common sense’, blanketly requiring written documentation of consent in all contexts. Van den Hoonaard (2001: 28–31) shows that insistence on formal written consent can sabotage its own goals, introducing suspicion, confusion and mistrust into the relationship between researchers and participants. Marshall (2006: 32–33) points out that these problems are magnified where consent forms need to be translated across languages and cultures, as Pettitt’s research (see Appendix 1) richly illustrates.

There is also on-going discussion about the related issue of when to document informed consent. Conventionally, it is a gateway exercise, completed once and for all before data are gathered. But Marshall (2006: 35), Bigelow and Pettitt (2015), Yeager-Woodhouse and Sivell (2006), among others, point out advantages to treating informed consent as a dynamic, on-going, process of communication that researcher(s) and participants revisit from time to time, to re-establish that participants’ involvement is voluntary throughout the course of a study. Re-conceptualizing informed consent as a dynamic process, however, entails that researchers accept that participants may at any time withdraw their consent, no matter how hard-won that consent was, or how disruptive withdrawal may be.

5 What are the consequences of ‘ethics creep’?

A final matter of debate about informed consent turns the problem on its head. Haggerty (2004) argues that the penetration of biomedical ethics into the social sciences has induced what he calls ‘ethics creep’, the process by which an ethics-regulatory bureaucracy expands to control more institutions and practices while simultaneously intensifies its surveillance of them. According to Haggerty, ethics creep distorts research in several ways with reference to informed consent: for ethnography in particular, the requirement for legalistic written documentation is ‘alien, unduly formal, and occasionally unworkable’ (p. 404); moreover, informed consent rules out certain ‘valuable forms of critical inquiry’ (p. 406) that cannot be undertaken without some masking of their purposes. Haggerty’s first objection might be addressed by some of the alternative means of documenting consent discussed above. His second objection reprises the long-sustained debate about deception in research.⁷ But Haggerty raises an important general issue in claiming that ethics creep signals ‘a move away from a system based on the assumption of professional competence and responsibility to one based on institutionalized distrust’ (p. 393). Van den Hoonaard (2001: 24) would seem to agree, in that he compares the proliferation of ethical constraints on the social sciences to a kind of ‘moral panic’, arising in response to widespread erosion of public trust: instead of helping researchers develop broad, sensitive, flexible, judgment, we fall back on formal regulation of their behavior.

For several reasons, scholars of L2 acquisition need to be attuned to this debate even as it extends beyond our immediate concern to get our projects approved and underway. A first reason is that one may be called to serve on an institutional IRB, and with that would be entrusted to participate in ethics review from a different angle. Another reason is that in the co-ordinate roles of teacher and administrator, scholars have opportunities to either help students develop their judgment, or simply impose rules on them; better yet, scholars have opportunities to model for students how they themselves work to ensure that their research is not only valid scientifically, but also satisfies both their personal values and meets external ethical standards, sometimes multiple communities’ ethical standards. A third reason is that research on L2 acquisition often invites researchers into relationships across cultural boundaries, opening insight which can enrich the debate about the role that informed consent has, or should have, in social-scientific research.

V Conclusions

The concept of informed consent as an essential constraint on research involving humans passed from biomedicine into the social sciences, where it has partially adapted to new disciplinary environments. Research on L2 acquisition has been slow to identify informed consent as a key professional responsibility. This is attested in textbooks, the websites of professional associations, and a survey of recent journal publications, even though L2 research is supported by institutions, governments, and funding sources committed to ethical codes in which informed consent is a cornerstone. In this review article, we have narrated an example of difficulties that can arise in L2 research sensitive to the demands of informed consent. Happily, scholars of L2 acquisition are not alone in wrestling with these complexities, as there is abundant on-going debate in the social sciences that addresses the problems that accompany more thorough incorporation of informed consent into research practices.

Footnotes

Appendix 1

Narrative illustrating some of the ethical challenges posed by research on L2 learning

Acknowledgements

Thomas thanks Doyle Calhoun for his outstanding research assistance with this text.

Declaration of conflicting interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Agre

Campbell

Goldman

. (2003) Improving informed consent: The medium is not the message. IRB: Ethics and Human Research 25: S11–S19.

Benitez

Deveaux

Dausset

(2002) Audiovisual documentation of oral consent: A new method of informed consent for illiterate populations. Lancet 359: 1406–07.

Bigelow

Pettitt

(2015) Narratives of ethical dilemmas in research with immigrants with limited formal schooling. In: De Costa

(ed.) Ethics in applied linguistics research: Language researcher narratives. New York: Routledge, pp. 66–82.

Blom

Unsworth

(eds) (2010) Experimental methods in language acquisition research. Amsterdam/Philadelphia, PA: John Benjamins.

Breese

Burman

Goldberg

Weis

(2007) Education level, primary language, and comprehension of informed consent process. Journal of Empirical Research on Human Research Ethics 2: 69–79.

Brown

Rodgers

(2002) Doing second language research. Oxford: Oxford University Press.

Clough

Campbell

Aliyeva

Mateo

Zarean

O’Donovan

(2013) Protocols for protection of human participants: A comparison of five countries. Journal of Empirical Research on Human Research Ethics 8: 2–11.

Coady

Bloch

(eds) (1996) Codes of ethics and the professions. Melbourne: Melbourne University.

Corrigan

(2003) Empty ethics: The problem with informed consent. Sociology of Health and Illness 25: 768–92.

10.

Dörnyei

(2007) Research methods in applied linguistics. Oxford: Oxford University Press.

11.

Faden

Beauchamp

(1986) A history and theory of informed consent. Oxford: Oxford University Press.

12.

Ghandour

Yasmina

El-Kak

(2013) Giving consent without getting informed: A cross-cultural issues in research ethics. Journal of Empirical Research on Human Research Ethics 8: 12–21.

13.

Gostin

(1995) Informed consent, cultural sensitivity, and respect for persons. Journal of the American Medical Society 274: 844–45.

14.

Haggerty

(2004) Ethics creep: Governing social science research in the name of ethics. Qualitative Sociology 27: 391–414.

15.

Hatch

Farhady

(1982) Research design and statistics for applied linguistics. Rowley, MA: Newbury House.

16.

Hatch

Lazaraton

(1991) The research manual: Design and statistics for applied linguistics. New York: Newbury House.

17.

Haverkamp

(2006) Ethical perspectives on qualitative research in applied psychology. Journal of Counseling Psychology 52: 146–55.

18.

Johnson

(1992) Approaches to research in second language learning. New York: Longman.

19.

Kaufman

Ramarao

(2005) Community confidentiality, consent, and the individual research process: Implications for demographic research. Population Research and Policy Review 24: 149–73.

20.

Koulouriotis

(2011) Ethical considerations in conducting research with non-native speakers of English. TESL Canada Journal / Revue TESL du Canada 5: 1–15.

21.

Kripalani

Bengtzen

Henderson

Jakobson

(2008) Clinical research in low-literacy populations: Using teach-back to assess comprehension of informed consent and privacy information. IRB: Ethics and Human Research 30: 13–19.

22.

Kubanyiova

(2008) Rethinking research ethics in contemporary applied linguistics: The tension between macroethical and microethical perspectives in situated research. Modern Language Journal 92: 503–18.

23.

Levine

(1991) Informed consent: Some challenges to the universal validity of the western model. Journal of Law, Medicine, and Ethics 19: 207–13.

24.

Mackey

Gass

(2005) Second language research: Methodology and design. Mahwah, NJ: Lawrence Erlbaum.

25.

Mackey

Gass

(eds) (2012) Research methods in second language acquisition: A practical guide. Chichester: Wiley-Blackwell.

26.

Marshall

(2006) Informed consent in international health research. Journal of Empirical Research on Human Research Ethics 1: 25–42.

27.

McNutt

L-A

Waltermaurer

Bednarczyk

. (2008) Are we misjudging how well informed consent forms are read? Journal of Empirical Research on Human Research Ethics 3: 89–97.

28.

Nunan

(1992) Research methods in language learning. Cambridge: Cambridge University Press.

29.

Paltridge

Phakiti

(eds) (2010) Continuum companion to research methods in applied linguistics. London: Continuum.

30.

Perry

(2011) Ethics, vulnerability, and speakers of other languages: How university IRBs (do not) speak to research involving refugee participants. Qualitative Inquiry 17: 899–912.

31.

Phakiti

(2014) Experimental research methods in language learning. London: Bloomsbury.

32.

Porte

(2002) Appraising research in second language learning. Amsterdam/Philadelphia, PA: John Benjamins. 2nd edition 2010.

33.

Reinharz

(1993) On becoming a social scientist. 4th edition. New Brunswick, NJ: Transaction.

34.

Robinson

(2010) Informed consent among analog people in a digital age. Language and Communication 30: 186–91.

35.

Seliger

Shohamy

(1989) Second language research methods. Oxford: Oxford University Press.

36.

Thomas

(2009) Ethical issues in the study of second language acquisition: Resources for researchers. Second Language Research 25: 493–511.

37.

van den Hoonaard

(2001) Is research-ethics review a moral panic? Canadian Review of Sociology / Revue canadienne de sociologie 38: 19–36.

38.

Walkup

Bock

(2009) What do prospective research participants want to know? What do they assume they know already? Journal of Empirical Research on Human Research Ethics 4: 59–63.

39.

Yeager-Woodhouse

Sivell

(2006) Prepackaged tour versus personal journey: The meaning of informed consent in the context of the teacher-study group. Journal of Academic Ethics 4: 189–203.