by Oliver Shoulson

Introduction

A little over a month ago, we members of the Yale class of 2020 started receiving our diplomas in the mail. For me and surely many others, this formal admittance to some kind of official degree or qualification came with the awkward consideration of what it means to call yourself an -ist. This year, 2020, is a painful and uncanny time to be anything, no less an infant physicist, classicist, psychologist, or linguist and no less due to the host of other identities with which these -ists may intersect. In the first ten weeks of my life as a (White ¹) linguist, I’ve had the chance to be — by the loosest of definitions — a professional linguist in working here at the YGDP. Here from the comfort and safety of my desk in my childhood bedroom, I set about creating and compiling research on linguistic diversity within the English spoken in North America, much of which focuses on the documentation and analysis of the language of minority groups, including African American Language.

Meanwhile, outside of my quarantine bubble, much of the conversation, tears, and agony occurring in this country has been rightly centered on the ongoing and systemic subjugation of Black bodies: Black bodies as the neck under the boot (or knee) of police and government, Black bodies as the target of desecration and mockery by the wearer of blackface. This phrase, ‘Black bodies,’ is, in its synecdoche, deliberately reductive and only one part of an implied duality (or multiplicity). It encapsulates, in part, the white “racist imago” of Black people, devoid in some crucial respect of humanity (Yancy 2015) ². Although it leaves deliberately undefined the missing component(s), it’s hard to sit here reading and writing about the equally ongoing marginalization and stigmatization of Black speech without joining some puzzle pieces, and experts far better suited to this discussion than I am have done the same (see Sharese King’s recent Op-Ed). The foremost of the issues we face this year—the global pandemic and massive nationwide reckoning with racist violence—are profoundly intertwined with how we listen to and talk about marginalized language. COVID-19 disproportionately ravages communities of color (CDC 2020; Oppel Jr. et al. 2020), and study after study has shown that Black patients’ complaints of symptoms and pain are more likely to be diminished or disregarded by healthcare professionals (Hoffman et al. 2016). Specifically, those complaining of COVID symptoms were six times less likely to receive testing or treatment (Rubix Life Sciences 2020). That Black speech is seen as unintelligent, not credible, or not meaningful cannot be disjoined from the capacity of a doctor to ignore complaints of illness or of four police officers to simultaneously disregard a painfully and resoundingly uttered: “I can’t breathe.”

The ostensibly neutral and informative work of YGDP phenomena pages that I’ve spent the last ten weeks on starts to feel blaringly insufficient in scope, and I return once again to the passage from Rickford above. Many publications on linguistic diversity and discrimination have begun with reference to the “unequal partnership” that he describes. As linguists, we study the languages of the world, their varieties and their manifestations with the overarching academic goal of enriching the wealth of knowledge pertaining to this miraculous entity, Language. The ever-growing and branching body of theoretical understanding is fed and watered by exactly the kind of descriptive and comparative work that we do here at the YGDP, and for that reason, minority language groups — often holding relatively little political and economic sway — constitute disproportionately large assets to the academic goals of those who study them. Within the study of the English language, the preeminent example is African American Language (AAL). I won’t list here, as Rickford begins to, the virtually endless ways that the study of African American speech has enhanced the field of linguistics — hopefully, in this essay, I will begin to convey a sense of these ways — but rather I will take this notion to its moral and logical conclusion: American linguists have a largely unfulfilled obligation to these marginalized speech communities in return.

On behalf of the YGDP, this essay is both a commitment to and an incremental stage in the ongoing fulfillment of that obligation. Acknowledging that there is no such thing as ideologically neutral description of language (Blake 2016: 155; Irvine & Gal 2000), we should be explicit about ours, and by articulating here in writing that our work at the YGDP is intended unreservedly to bolster and empower marginalized dialects, their legitimization, and the acknowledgement of their complexity and social significance, we take a crucial step in turning research on into research on and for (Cameron 1992: 22). The final stage: research on, for, and with is also a work in progress in many Linguistics departments and academic communities more broadly, and something I take up further at the end of this essay. My goal, therefore, is not just to reiterate some of what YGDP phenomena pages and some primary research say pertaining to AAL so as to raise awareness of and dispel various myths and biases (though there will be some of that), but to expound overtly on why it’s important to learn about and legitimize grammatical diversity, including the many material consequences of not doing so.

Finally, I should note that this essay is aimed at much the same audience as our phenomena-pages and FAQs; that is, people interested in language and linguistic diversity but without an extensive background in the field of linguistics. It is by no means exhaustive or categorically authoritative, and many of the topics addressed below are subject to ongoing research. The reason I have structured this piece something like an academic article is to have the opportunity to cite and direct readers to some of the brilliant and cutting-edge primary literature on these topics (particularly from the last few years), so let this essay be your appetizer, not your entree. With that, we begin:

Grammatical Diversity and Group Affiliation

Many people are already familiar with the notion of a shibboleth. The concept derives from the Biblical story in Judges 12 which I’ll include here in the JPS translation:

The Gileadites held the fords of the Jordan against the Ephraimites. And when any fugitive from Ephraim said, “Let me cross,” the men of Gilead would ask him, “Are you an Ephraimite?”; if he said “No,” they would say to him, “Then say shibboleth”; but he would say “sibboleth,” not being able to pronounce it correctly. Thereupon they would seize him and slay him by the fords of the Jordan. Forty-two thousand Ephraimites fell at that time.

Judges 12: 5-6

The term shibboleth now has little to do with its original meaning of ‘the head of a stalk of grain,’ but rather refers more generally to some kind of linguistic element used to identify oneself (or to diagnostically identify another person) with a particular group. Often when we think of shibboleths, our minds first go to phonological or lexical phenomena; that is, to linguistic matters concerning how words sound (like in the Biblical story) or the choice of particular words over others for a certain purpose.

One example of a phonological shibboleth for North American English pertains to what linguists call the pin/pen merger. If someone pronounces the words pin and pen pretty much the same (i.e. there’s little difference in the central vowel), they likely grew up speaking a Southern dialect of American English. If someone pronounces the a in class just as tight and nasal as the a in clam, they likely grew up around Chicago, Cleveland, Detroit, another city in that region. Likewise, many of us already have a knowledge or intuition of certain lexical shibboleths for American English. If someone prefers the word pop to refer to a fizzy beverage as opposed to soda they’re likely from the Midwest. If a speaker can use the word wicked as an adverb in addition to an adjective they’re likely from New England.

The YGDP is concerned with a third category of variation pertaining neither to word-choice nor word pronunciation, but rather how words are combined, ordered, and altered for syntactic function; we call these grammatical variants. Many of these are just like the phonological and lexical examples above in that they are mainly determined by geographical distribution and have little social stigma associated with them one way or another. For instance, if you grew up in certain areas of the American Midland (say, Pittsburgh) you might be able to say something like: the car needs washed (see the needs washed page). To non-Midland speakers, it sounds like there’s something missing or wrong about that sentence. Did you mean the car needs to be washed? Or the car needs washing? Nope; both of those constructions are fine too, but Midland speakers just have an additional way of expressing that same sentiment.

Grammar does, of course, vary along axes other than geographic region. Nearly any sociological factor you can think of (age, gender, education, profession, religion, race etc.) has the potential to interact with grammar in some way. Sociolinguistics is constantly juggling the unique idiosyncrasies of every person’s individual language alongside the need to define particular language groups and the language that they use. It’s not necessarily intuitive to think of people we see around us every day, take classes with, hang out with, or work alongside as having a different language, but it’s an important notion to consider:

African American Language

Whether or not you consider African American speech a language obviously has more to do with your definition of language than whether or not you believe African Americans tend to speak differently than other American groups. The question of what counts as a language is interminably complex, but it is also entirely inseparable from geopolitics. Linguists are fond of reciting the famous Max Weinreich quote: “a language is a dialect with an army and a navy.” That is to say: language boundaries are — in their folk/popular conceptions — often drawn along the same borders established by the political authorities bearing their names or political groups seeking to identify as such (see e.g. Puzey 2012). The conflation of political regions with linguistic regions goes both ways when it comes to mislabeling. Mutually intelligible speech may be labelled as two distinct languages because the speakers fall within different borders — this is the case with Danish and Norwegian — and mutually unintelligible dialects may be grouped beneath one umbrella by virtue of their shared sovereign government — as with Mandarin and Cantonese both being called Chinese. You may have heard terms which imply that African American language is a dialect of English (it is commonly referred to as African American (Vernacular) English — AA(V)E — for this reason), and there are plenty of experts who would classify it exactly this way; however, there are several reasons why African American Language is catching on as a term in place of AA(V)E, Ebonics, etc.:

1. It combats the harmful urge to group African Americans into a linguistic monolith; for the general public, “dialect” may feel instinctively like an atomic unit of language variation, and terms like AA(V)E have the potential to undersell the linguistic variation within black speech communities (see e.g. King 2020; Fasold 2019). This overgeneralization is harmful because it has the potential to bolster other inaccurate generalizations about a racial group, and assuming linguistic uniformity across wide swaths of people and regions can actually have dire real-world consequences (see the discussion of LADO below).
2. Some—though certainly not all—historical linguists feel it is historically inaccurate to refer to AAL as a variety of English as opposed to its own linguistic entity (i.e. a creole or contact language); more on this below.

The fact remains, however, that children who grow up speaking AAL at home and Mainstream American English (MAE) in the classroom acquire a kind of functional bilingualism and the ability to rapidly switch between the languages. This carries with it all of the cognitive benefits typically associated with bilingualism including advances in executive function and critical thinking skills (Lee-James & Washington 2018; Fricke & Kootstra 2016).

AAL Grammar and Systematicity

That AAL is ‘bad grammar,’ ‘grammarless,’ or ‘bad English’ is a widely held belief across racial lines; you’ll find many native and fluent speakers of AAL who nonetheless believe that their mother tongue is grammarless or uses poor grammar (Fasold 2019: 205). This notion is harmful because it reliably goes hand in hand with a value judgment about AAL and, by hardly any extension, African American people. The moment AAL is viewed as grammatically inferior, people feel justified in considering its speakers less articulate and intelligent. When a linguist talks about grammar, they refer to the systematic rules which anyone fluent in a language has internalized, allowing them to construct and interpret meaningful sentences. These rules may include some of the ones you learned in elementary school, and may not include others, but what’s most important is that they are systematic (not random). AAL is unquestionably and indisputably governed by systematic grammatical rules, even if they differ in many ways from those of MAE. Let’s take a brief look at a prominent grammatical feature of AAL which distinguishes it from MAE: null copula. This is the omission of a form of the verb to be in the present tense as in:

Some of them __ big. Some of them __ small.
(Green 2002: 52)

This is a widespread phenomenon in AAL and, in many languages, it is a feature of the mainstream grammar taught to children in schools. You’ll hear me cite Hebrew (Biblical and Israeli) a lot because it’s one of the few languages in which I’m proficient, but see the following equivalent sentence:

כמה

kama

some

מהם

m-hem

of-them

גדולים

g'dolim

big

כמה

kama

some

מהם

m-hem

of-them

קטנים

k'tanim

small

'Some of them (are) big. Some of them (are) small.'

There’s nothing “bad” about this Hebrew sentence, it’s perfectly grammatical to any Modern Hebrew speaker. The notion that a similar construction in AAL is “bad” is rooted simply in the entirely arbitrary notion that black people should speak like the overwhelmingly white mainstream. An unfamiliar observer might mistakenly conclude that AAL-speakers are just dropping be-verbs willy-nilly, but, in fact, the omission of copula is highly systematized. We already pointed out that null copula generally only appears in the present tense; additionally, a copula can never be omitted at the end of a clause. That is, you are very unlikely to hear an AAL speaker say:

*I don’t care what you __.
(Labov 1969: 721)

in order to mean: ‘I don’t care what you are.’

Additionally, a copula cannot be omitted for first person subjects. You’re unlikely to hear an AAL speaker say:

*I __ driving to Amherst.
(Green 2002: 40)

in order to mean: ‘I’m driving to Amherst.’

See the YGDP page on null copula for further systematic analysis. All of this is to say that those linguistic properties which may look “lazy” or “random” to the untrained observer are not lazy or random at all, but rather linguistic properties which AAL shares with many languages, just not MAE.

Finally, it’s always fun to point out the various ways in which AAL’s grammar allows its speakers to make much finer and more nuanced distinctions of meaning for which MAE would have to resort to lexical distinctions or greater verbosity. AAL has a number of function words for marking aspectual distinctions on predicates (that is, distinctions involving the duration and completion of an event) for which MAE has no standard grammatical process. These include stressed BIN, habitual be, and aspectual done. Thus, in the following sentence, an AAL speaker can express the same sentiment using specialized grammar that it would take an MAE speaker a certain amount of linguistic acrobatics to express:

AAE: They just sent me this one, but I BIN having that one.
MAE: They just sent me this one, but I have had that one for a long time.
(Green 2002: 56)

AAL Logic and Coherence

Another common trope in the discrimination against black speech is the notion that AAL is illogical or incoherent. This misconception derives entirely from the misapplication of one language’s syntactic/semantic principles (i.e. MAE’s) onto another. While the linguistic history of AAL may, in part, be rooted in attempts by white people to render black people incoherent (African slaves from different linguistic groups were often assigned to work together to prevent uprisings — an idea as old as Genesis 11, McElhinney 2019: 168), the miraculous coherence of the language faculty prevails. One phenomenon often ridiculed as “illogical” in AAL is called negative concord exemplified in the following:

Nothing don't come to a sleeper but a dream.
(see the YGDP page)

I’m sure we’ve all heard someone say that a sentence like the above contains a “double negative” and thus “actually means” the opposite of what was intended. The fact is, once again, the grammatical principle of negative concord is shared by hundreds of languages around the world. Returning to Modern Israeli Hebrew, a direct translation of the above sentence would be the following:

שום–דבר

shum-davar

nothing

לא

lo

not

בא

ba

come

אל

el

unto

היושן

ha-yoshen

the-sleeper

אלא

ela

except

חלום

xalom

dream

'Nothing (*doesn't) comes to a sleeper but a dream'

Note the “double negative.” In many languages, negative elements in a sentence have to agree with each other, just as in English, a subject and verb have to agree in person and number. Logically speaking, there’s no reason why you have to mark both the subject and a verb of a sentence as singular; one should be sufficient:

He

3MS

likes

like.3S

green apples

...

Likewise, linguists argue that nothing and don’t are simply agreeing (in a sense) in negativity, and that’s why we call it “negative concord” and not “double negatives,” just like we call it “number agreement” and not “double singular.”

It’s actually already a misconception to associate this English negative concord construction exclusively with AAL. Negative concord is an incredibly widespread feature of nonstandard Englishes, and it is many hundreds of years old (Nevalainen 2006). The impression that this phenomenon, unnatural to many, is incorrect or nonsensical is a reflection of the incorrectness illusion — the oft-observed illusory and misguided notion that an unfamiliar construction is incorrect (Horn 2014: 342). Likewise, the impression that unfamiliar phenomena are “newfangled slang” is a reflection of the recency illusion — the illusory and misguided notion that an unfamiliar construction must be new (Zwicky 2005).

You may find AAL incoherent because it may not be immediately intelligible to you, not because it is unintelligible.

AAL Intelligibility and Credibility

For this section, I turn largely to the Hall-Lew et al. (2019) essay which takes up the question of intelligibility vs. credibility directly. There is an innocent enough misconception — likely held by many across racial lines — that the dearth of credibility afforded to AAL speech is due, at least in part, to its not being understood. This assumption is wrong, and its harm is that it provides a convenient and, ultimately, fallacious excuse for people to disbelieve marginalized speech when in fact, intelligibility and credibility are entirely orthogonal properties.

This was shown by Hall-Lew et al. in a comparative study of AAL, for which lower intelligibility is associated with lower credibility, and the Scots-English of Scottish heritage tour guides, for whom lower intelligibility is associated with higher credibility. Their experiment showed that for non-Scottish tourists, a tour guide with a thicker and therefore less intelligible Scottish dialect was perceived as more authentic to the region and thus more authoritative and trustworthy.

There is little to say about this myth in terms of grammatical diversity other than that your impression of a speaker’s credibility likely comes from your own predisposition for bias against a particular language community rather than from your difficulty understanding their speech. This phenomenon is nowhere better illustrated than the groundbreaking Rickford and King (2016) paper regarding Rachel Jeantel’s testimony in the trial of George Zimmerman for the killing of Trayvon Martin. Jeantel exhibits the typical cognitive dexterity of a multilingual speaker, rapidly code-switching between AAL, Afro-Jamaican, and MAE constructions (see for example, her response to the court stenographer’s request for clarification on page 956 of Rickford and King, wherein she makes a copula overt that had previously been null); she is also a fluent speaker of Spanish. Nonetheless, her speech was deemed unintelligent and untrustworthy; although her testimony pertained to some of the most crucial facts of the trial — whether Martin was running toward or away from his killer at the time of the shooting — it hardly even came up in jury deliberations. Rickford and King show how specific elements of her non-white linguistic repertoire were used to dismantle perceptions of her intelligence and credibility.

Lastly, it’s important to note that perceived speaker intelligibility is not absolute or immune to hearer bias on its own. In a 2015 study (Babel and Russel), which reproduced findings from the groundbreaking Rubin & Smith (1990) study, it was shown that when primed with a photograph of a Chinese Canadian person, White Canadians judged identical voice recordings as more accented and less intelligible than when primed with a photo of a White Canadian person. This phenomenon has been popularly dubbed “accent hallucination” and it is another reason why intelligibility is a deeply flawed basis for the assessment of speaker credibility.

Diversity within Diversity

In the early days of its study and documentation, AA(V)E was defined exclusively around the speech of young black men from the inner city (hence “Vernacular”)(Eckert 2003). The fact is, African American speech is far more diverse than that, and speakers are likely to draw from multiple linguistic repertoires in order to index the nuanced and unique intersections of their identities. Benor’s (2016) study takes up the linguistic behavior of several Black Jews, who draw variably on AA(V)E and Jewish English linguistic elements in order to encode their sociolinguistic alignment in a given context. Conversely, non-black speakers are constantly adopting and appropriating linguistic elements of AA(V)E; many such grammatical variants are documented on the YGDP website as having originated in African American speech and spreading into general parlance; these include you do you, fixin' to, tryna and many more.

The recognition and documentation of intra-group linguistic diversity in general is absolutely imperative for a variety of pressing reasons. In researching this essay, I came across a phenomenon entirely new and utterly appalling to me that has been termed Language Analysis for the Determination of Origin (LADO). As the name suggests, this procedure — which is heavily in use in the United Kingdom and elsewhere — involves interviewing refugees purely for the purpose of ascertaining, based on their language, where they come from and, thus, whether or not they are eligible for political asylum. The actual analysis is conducted by ostensible linguistic experts, though not necessarily with a specialty in the language spoken by the refugee in question. Patrick et al.’s (2019) book describes this procedure, its application, and myriad cases in which it has gone horribly wrong. Although the procedure itself is in dire need of reform and, likely, elimination, the existence of such forensic linguistic techniques make the ongoing work of detailed, quantitative, and unbiased sociolinguistic research (lexical, phonological, and grammatical) absolutely crucial. Assumptions made about speakers’ origins and group alignment based on their speech must be highly informed or disregarded entirely, and the prevalence of misinformation is something that many early and well-intentioned AA(V)E scholars and studies inadvertently contributed to.

AAL Origin Stories

At this point, I thought it would be useful to address briefly a question that may be on some people’s minds, namely: where does AAL come from and how did it develop? Like all of the other issues in this essay, these questions have been the subject of much debate and no less historical consequence. In 1996, the Oakland Unified School District in California passed what became known as the Ebonics Resolution, which proclaimed that Ebonics — their term for AA(V)E — was an African language and not a variety of English. This was done with the noblest of intentions. Among other things, the school board thought that children growing up speaking mainly Ebonics should have access to English language enrichment alongside designated Ebonics education just like other bilingual students have access to heritage language programs and English as a Second Language enrichment. The resolution specifically refers to Ebonics as “such West and Niger-Congo African language” and also makes reference to what they call “Pan-African Communication Behaviors” (Oakland Unified School District, Office of the Board of Education 1998).

The notion that AAL is typologically and phylogenetically an African language has been touted at various points by some politicians and Afrocentrist activists. In the linguistics community, its strongest analog is called the the Creolist Hypothesis, which posits that AAL is fundamentally a creole (i.e. a language resulting from contact between linguistically disparate groups) of African/Caribbean languages and English; this would make it a typologically distinct entity from English. There is nothing inherently wrong with this idea, but the truth is that there is little consensus in the linguistics community about AAL’s origins, and really, the notion that in order to achieve political legitimacy, AAL must be proclaimed a direct offshoot of Niger-Congo languages is the whole problem. Just because AAL is probably not an African language does not mean that speakers of both AAL and MAE are not, in some sense, bilingual. It also does not mean that AAL cannot be called its own language for sociological and ideological reasons. The whole question of origins is, in most cases, a red herring.

That said, for those interested, some of the AAL origin theories that are most prominent in the linguistics community now are the Neo-Anglicist Hypothesis and the Substrate Hypothesis (see e.g. Veenendaal, Straatjes & Zeijlstra 2014, Hutcheson and Cullinan 2017, Green 2002 ). The former posits that AAL is a collection of dialects descended from Englishes brought over by early British settlers, that it shares a common ancestor with Southern White Vernacular English (SWVE), and that it has since undergone divergent language evolution like many languages do. The latter claims that AAL was born out of contact between early settler dialects and some already-existing English-based creoles (perhaps you’ve heard of Gullah, an English creole of South Carolina, Georgia, and neighboring islands). Understanding the ins-and-outs of these hypotheses is not necessarily crucial; just know that AAL’s history is still shrouded in mystery, and it is an exciting research avenue on the forefront of many linguists’ agendas.

Conclusions

As I mentioned in the introduction, the final stage in Cameron’s progression: research on, for, and with, takes much more than an essay to remedy. Involving and partnering with marginalized speech communities in linguistics is a multifaceted endeavor. It may look incorporating joint study/service regimens into linguistics curricula that examine marginalized speech and linguistic diversity. In the cases of more insular and endangered language communities, it may take the form of the strategy now employed in many indigenous language documentation and revitalization projects wherein outside linguists take a much more hands-off approach, deferring to the expertise and priorities of those within the speech communities (see e.g. Cooper 2019, Bowern 2008). It also undoubtedly looks like making the academic environment in which we train linguists more welcoming to members of those speech communities. That is certainly not to say that a Black faculty member or student should specialize in Black speech, but failing to represent Black speakers in the linguistics departments which study them does little to counteract the ‘othering’ of AAL. It would be misguided to assume that the institutional racism resulting in the widespread disregard for Black language is at all distinct from the institutional racism resulting in Black scholars’ underrepresentation in fields like linguistics. That the present author — a recent graduate who happened to be working for the YGDP when the country tipped into the current peak of outrage — is a White man is no coincidence. In fact, there was not a single black person in my graduating linguistics class. In 2014, Black people made up roughly 12% of the US population, roughly 15% of undergraduate enrollment, and only 4% of Bachelors degrees in linguistics nationwide (Espinosa et al. 2019, National Science Foundation 2017).

One of the most wonderful things about the work of John Rickford, whose quotation introduced this essay, is that his scholarship is often accompanied by tangible action items for linguists to take in promoting linguistic social justice and giving back to the communities that form the bedrock of their scholarship. Among these are ideas for classes that couple the study with on-the-ground community work as mentioned above and various ways for Linguistics departments to promote diversity and inclusion in their scholarship. I think it’s hardly controversial to say that this advice is something we in higher education should take to heart. This coming year, the department is forming a Committee on Diversity and Inclusion with representatives from the undergraduate and graduate programs alongside faculty, and the YGDP has and will continue to manifest its research and outreach through on-campus advocacy including Linguistic Prejudice Workshops at the Center for Teaching and Learning, film screenings, etc. In the end, though, it’s on us linguists (including you, if you’ve read this far; that’s another -ist for you) to disabuse people of the harmful assumptions and misconceptions that lead to the stifling of Black speech in addition to that of Black Lives in general. Turning to the first-person-singular, I also feel compelled to say that it is on all of us regardless of our respective -ists to devote our time and money to these issues. Distributing knowledge is valuable; as is distributing resources. Time for me to get up from this desk and get going again.

Notes

1. In accordance with present custom, I capitalized ‘Black’ throughout the essay. There is ongoing debate as to whether or not ‘White’ warrants the same treatment. One argument made in the Center for the Study of Social Policy’s recent post is that to capitalize ‘Black’ and not ‘white’ is to define Blackness with respect to an implicit white norm, which I certainly want to avoid. On the other hand, capitalizing ‘White’ has the potential to evoke the rhetorical and stylistic conventions of white supremacist discourse. I’m not trying to take a side one way or another on this, but I’d be remiss not to mention it.
2. I should also note that this phrase has particular historical significance in discussion of trans black people due to the cultural fixation on their bodies and the resultant encroachment on their safety and autonomy.

References