14.2.1 “Cribbed” Words and Polyvocality

The consequences of dictionaries’ long associations with manners and morality can be heard in the responses to the Third discussed in the introduction. Before tracing the origins of those associations to early English-language dictionaries, however, we begin with a brief look at the polyvocal nature of even those first monolingual dictionaries. Their multivalent properties arise partly because most, on some level, are collective enterprises. As has been long recognized, early dictionaries, though usually attributed to single authors, are often built upon the previous work of others. It is a practice that N. E. Reference OsseltonOsselton (1958, 143) calls the “cribbing-method” of dictionary building, which he argues led to a kind of uncritical replication of entries “regardless of usage or custom” in the seventeenth and eighteenth centuries. Edward Phillips, for example, famously borrowed freely from Thomas Blount’s Glossographia (1656) for his New World of English Words (1658) without acknowledgment, leading Blount to respond with his A World of Errors Discovered in the New World of Words (1673). A rough accounting of entries in New World of English Words with similar, corresponding entries in Glossographia identifies 218 potentially cribbed candidates – or 1.78 percent of all entries in New World of English Words (see Figure 14.1).Footnote ²

Figure 14.1 Heatmap showing locations of entries in the New World of English Words possibly cribbed from Glossographia.

While Figure 14.1 is certainly suggestive in highlighting Phillips’ potential cribbing from Blount, the entries it identifies are not definitively lifted from Glossographia. One complicating factor is that Blount, himself, was a prodigious cribber. Glossographia borrows from the work of John Reference BullokarBullokar (1616) and others, which Blount references in his frontmatter. Unacknowledged, however, are his borrowings from the Latin–English dictionaries of Reference ThomasThomas (1587) and Reference HolyoakeHolyoake (1639), which Starnes and Noyes document (Reference Starnes, Witt and Noyes1946, 40–42). Those cribbed entries are a contributing factor to Glossographia’s polyvocality, though not the only one. In his frontmatter, Blount expresses acceptance of, if not admiration for, some kinds of language change. English has been “enobled,” he observes, “by admitting and naturalizing thousands of forein Words, providently brought home from the Greek, Roman and French Oratories; which though, in the untravel’d ears of our Fathers, would have sounded harsh.” Immediately following this preface is a poem attributed to J.S. (presumably his friend John Sergeant). In contrast to Blount’s seemingly more sanguine view of the transformations attending seventeenth-century English, the poem invokes the story of Babel, opining that once “Language and Laws had firmly held together” but have since “degenerate[d] from their Mother.” It proceeds to map those anxieties onto English:

As Reference VișanRuxandra Vişan (2018) notes, the poem fits into a complaint tradition, which views language as being in a perpetual state of corruption and decline – thus threatening the social, cultural, and moral order. The role of the dictionary, according to this declinist vision, is to serve as a bulwark against the forces of chaos. Of Blount’s dictionary, the poet J.S. approvingly asserts that it both imposes order (“Confusion, in this Book, in Order’s set”) and, in so doing, reverses previous deterioration (“Old Babels Ruins this in part repairs”). The poem, therefore, sits in an interesting tension with Blount’s own self-professed curiosity regarding lexical innovations related not only to scripture, but also to trades like haberdashery, winemaking, and cooking, as well as to fields like anatomy, physics, and the law.

The tension between Blount’s stated enthusiasms and J.S.’s declinist foreboding highlights one of the many ways in which dictionaries can assert inconsistent, even conflicting beliefs about language and attitudes toward its speakers. Again, this may not be surprising given the myriad types of information organized in dictionaries. The examples from Glossographia point to the convention of advancing theories of language and linguistic history in the prefaces of early monolingual dictionaries; these can be rich sites from which to glean understandings of how the dictionary is imagined to fit into those narratives (Reference Álvarez, van Ostade and der WurffÁlvarez 2009). However, one should be wary of accepting frontmatter as the “true” or complete description of a dictionary’s ideological underpinnings (Reference AdamsAdams 2020; Reference Coleman and OgilvieColeman and Ogilvie 2009). There is insight to be gained from all of its contents, including a fundamental accounting of what words are there and what ones are not. To be sure, the narrow scope of hard-word dictionaries like Glossographia is partly the product of the limitations of human labor. Their focus, however, also reflects the era’s anxieties as the English lexicon swelled to meet new communicative demands brought on by its application in areas previously reserved for languages like Latin.

14.2.2 “Hard Words” and the Anxieties about Language Change

Early monolingual dictionaries from Robert Cawdrey’s A Table Alphabeticall (1604) to Nathaniel Bailey’s An Universal Etymological English Dictionary (1721) were meant to guide readers, particularly as they encountered words derived from Latin, Greek, and other foreign languages. Like J.S. (the writer of the poem prefacing Blount’s Glossographia) many creators of hard-word dictionaries believed that such guidance served not only a functional purpose but also a moral one. In Cawdrey’s case, those motivations are not necessarily transparent. On the one hand, he warns against “strange ynckhorne termes” and admonishes those who “seek so far for outlandish English, that they forget altogether their mothers language” – an endorsement of supposed native English wordstock, which has echoes in “plain English” movements into the present. On the other, his entire project purports to demystify those “strange ynckhorne termes” for his audience. This ambivalence, Sylvia Reference BrownBrown (1996) explains, can be attributed to his larger evangelical cause. As a minister and a Puritan, Cawdrey believed in an individual’s personal relationship with God, a relationship partly founded on and mediated by one’s familiarity with scripture. Thus, while he may inveigh against “counterfayting the Kings English,” his ultimate goal was to expedite his readers’ access to the word of God, whether that word was articulated in “the plainest and best kinde of speach” or not. This goal is gestured at in the subtitle’s reference to “Scriptures” and “Sermons” and suggested further by the fact that roughly one out of every six of Cawdrey’s headwords appears in The Book of Common Prayer. These include religious terms like ecclesiasticall, ministration, and sanctifie, as well as more general ones like communicate, expedition, and operation. The substantive overlap between the works is particularly suggestive given that, prior to writing his dictionary, Cawdrey was defrocked largely due to accusations that he took liberties with The Book of Common Prayer in his ministry. At his tribunal, “he was charged, not only with nonconformity, but want of learning” (Reference BrookBrook 1813, 439). Thus, while Cawdrey expressed strong opinions about the relative social and communicative value of lexical variants, he did not create his dictionary to be a catalog of preferred word forms or a bulwark against “lesser” ones. His conscripting of lexicography to the cause of evangelism was done in service of access – in some sense to puncture the elitism of religious gatekeepers embodied by their Latinate vocabulary.

Other hard-word dictionaries were more explicit in their moral imperatives. For example, in his dictionary, New World of English Words, Edward Phillips writes of his tempered enthusiasm for the changes affecting the English lexicon in the seventeenth century, with equivocation similar to Cawdrey’s. He begins by praising Edmund Spenser before identifying “his frequent use of obsolete expressions” as “the greatest blemish to his Poem, otherwise most excellent, it being an equal vice to adhere obstinately to old words, as fondly to affect new ones” (1658, c). In order to balance these competing “vices” – obstinacy and affectation, archaism and neologism – Phillips goes on to describe how he will “set [his] mark” upon those words that he finds lacking in purity, authenticity, or “a natural and unaffected stile” (1658, c2). Much like Cawdrey, Phillips clearly sets out to demystify the burgeoning store of loanwords for his audience (so that they will not be “amazed as if they had met with a Hobgoblin” when “they spy but a hard word,” as he puts it). Unlike Cawdrey, however, he wants to discriminate among the words he catalogs. In doing so, he invokes the language of morality, not only in characterizing some word choices as a “vice,” but also in formulating a binary that pits authentic, legitimate, and natural words against barbarous, spurious, impure ones. It is a formulation that has echoes in the responses to Webster’s Third quoted in the introduction and, as Anne Reference CurzanCurzan (2014, 101) observes, informs attitudes toward standard language and prescriptivism into the present.

14.2.3 The Domestication of “Hard Words” and the Policing of Gender Roles

Among early monolingual dictionaries, perhaps none is as clear in its statement of moral purpose as John Dunton’s The Ladies Dictionary (1694; Reference Considine and BrownConsidine and Brown 2010). In his dedication, Dunton (under the pseudonym N. H.) writes:

Many of the dictionary’s 1,941 headwords are biographical – names of women from religion, history, and myth, who are meant to serve as models (and sometimes cautionary counterexamples) of “Female Fortitude and Valour.” A variety of entries relate to courtship (courtship of men, love, kissing), marriage (conjugal state, husband, nuptial doweries), domestic life (cook-maids, chamber-maids, keeping house), appearance (body the beautifying thereof, fashions, hair), and conduct (confession of a leud woman, harlot, obedience of virgins). As both the above quotation and these examples of headwords suggest, the dictionary is largely concerned with the patriarchal policing of women’s bodies and behavior. Take, for example, the entry for pride which begins:

Dunton agonizes over conformity to normative gender roles, worrying that men have become too “effeminate” and women too “man-like.” He goes on to criticize both women’s comportment and dress as being “immodest.” The latter he fears attracts men’s “darting eyes.” As for the former, in the same entry he later opines that women’s talk has become “lascivious” and “so much and loud, that a few Women would suffice to make the Noise of a Mill” (401).

In dictionaries like Phillips’ New World of English Words, beliefs about speakers and communities are mapped onto language, thus turning words into indexes for those beliefs. Dunton does something rather different. He marshals the conventions of the dictionary in creating an inventory of sexual mores, social manners, domestic hierarchies, and in all spheres – sexual, social, and domestic – rationalizations for male dominance. Although The Ladies Dictionary is an extreme instance of how the definitional content of entries can communicate ideological positions, other dictionaries have done the same, albeit in usually less elaborated and overt ways. (The Hobson-Jobson dictionary is a good example, which is discussed in a later section.)

In addition, Dunton’s dictionary is not alone in imagining an audience made up of, at least partly, women. Cawdrey, for example, asserts that his words were “gathered for the benefit & helpe of Ladies, Gentlewomen, or any other vnskilfull persons.” Similarly, Henry Reference CockeramCockeram (1623) identifies his audience as “ladies and gentlewomen, young schollers, clarkes, merchants, as also strangers of any nation” and Blount “the more-knowing Women, and less-learned Men,” as well as “the illiterate, who can but finde, in an Alphabet, the word they understand not.” Sylvia Reference BrownBrown (1996) argues that this targeting of women (at least women of a certain class) as an audience suggests their equivocal place in an imagined linguistic and social order: as untainted by the knowledge of foreign tongues (and thus more “pure”), yet also vulnerable to linguistic and moral excesses due to their unschooled ignorance. The framing of women as at once pure, vulnerable, and dangerous suffuses The Ladies Dictionary. In the context of hard-word dictionaries, their perilous yet redeemable status makes women ideal (and idealized) conscripts in a project “to domesticate foreign words and, at the same time, to domesticate and propagate a godly programme of reformation” (Reference BrownBrown 1996, 361).

Related to its imagining of a gendered readership is the dictionary’s participation in the seventeenth century construction of heteronormativity. This is partly a discursive effect of Dunton’s emphasis on the moral implications of women’s romantic and domestic relationships to men, thus assuming a fixed binary. In other ways, however, it is made explicit, particularly in the entries The Ladies Dictionary copies directly from Glossographia. These include entries for terms at the intersection of law, sex, and religion such as buggery, sodomy, divorce, and bigamy. Consider the entry for buggery (which in Blount is spelled buggerie):

Obscuring physical specifics, the definition instead emphasizes the participants, who are sequentially joined by the preposition with, effectively pathologizing – by religion, law, and custom – any non-heteronormative union. Definitions of this kind (especially for buggery and sodomy) are common in early dictionaries, according to Stephen Reference TurtonTurton (2019). In addition, the entry’s use of Latin to conceal the most detailed part of the definition (“carnalis copula contra naturam, & hæc vel per confusionem Specierum, sc. […] vel sexuum”), while at odds with the stated purposes of hard-word dictionaries like Blount’s, is in keeping with a general tendency for dictionaries to mask sexual content or to avoid taboo words altogether (Reference Mugglestone and GorjiMugglestone 2007; Reference ReadRead 1934a).

Dunton borrows from other dictionaries besides Blount’s, however, and his cribbing from Stephen Reference BlankaartBlankaart (1693) in particular creates a tension not only with Blount but also with Dunton’s policing of sex, bodies, and women’s autonomy more generally. Blankaart, a Dutch physician, published an English-language edition of his Lexicon Medicum Graeco-Latinum in 1684, titled A Physical Dictionary. As he does with Glossographia, Dunton directly copies a number of entries from A Physical Dictionary. These include entries related to anatomy and medicine like catamenia, uterus, and umbilicus, with a particular focus on terms related to reproduction, which Dunton calls “the chief end of Marriage.” Whereas the entries cribbed from Blount are largely ideologically aligned with the rest of The Ladies Dictionary, those cribbed from Blankaart have a thoroughly different orientation toward gender and sexuality – one that is more medical than religious, more descriptive than dogmatic. Thus, just as Glossographia and its intertextuality illustrate the ideological tensions that can be present in dictionaries, its recycling in The Ladies Dictionary underlines the ways in which moral authority can be both reinscribed and contravened, sometimes in the same, polyvocal text.

14.3.1 The Transfiguration and Appropriation of “Foreign” Words

If the project of early hard-word dictionaries can be summarized as one of domestication, that of colonial dictionaries would be one of recontextualization. Published largely during the era of European colonial conquest from the late eighteenth through nineteenth centuries, these glossaries and dictionaries were part of a textual apparatus that classified and codified knowledge for the purposes of justifying and maintaining imperial rule. That apparatus included studies of local and historical languages (like Sanskrit), statistical reports, encyclopedias, and legal codes (Reference CohnCohn 1996; Reference ErringtonErrington 2008). It also included fictional representations of colonial subjects – representations that often drew directly from glossaries and reports published in magazines and elsewhere for their imagined depictions of speakers and their speech (Reference BrownBrown 2018). That colonial glossaries and dictionaries were created with very different goals than hard-word dictionaries is evident in the stated purposes to glossaries like the Glossary to the Fifth Report (Reference WilkinsWilkins 1813) and A Glossary of Judicial and Revenue Terms, and of Useful Words Occurring in Official Documents Relating to the Administration of the Government of British India (Reference WilsonWilson 1855). The first was appended to the Fifth Report – a review of the East India Company’s affairs commissioned by the British Parliament and published in 1812. It was prepared by Charles Wilkins and submitted one year after the release of the Fifth Report, specifically to assist government officials in England with the non-English words present in the documents. In his preface, Wilkins (a translator and printer employed by the East India Company, as well as a co-founder, with William Jones, of the Asiatic Society) complains about a lack of consistent rules for transliterating local scripts, when such words appear in English-language documents – a situation he attributes both to “our imperfect alphabet” and “the ignorance or inattention of the Native clerks” (Reference Wilkins1813, iii).

A Glossary of Judicial and Revenue Terms was similarly produced by a functionary of the East India Company, with the intent of assisting English speakers in decoding a variety of colonial documents rather than a single, specific one. Much like Wilkins, its author, Horace Hayman Wilson, writes of his dismay at and dissatisfaction with the anglicization of Indian orthographies. Wilson (the librarian to the company and a professor of Sanskrit at Oxford) characterizes local words when spelled with the English alphabet as taking on a “preposterous form” and as “so strangely disguised” when transliterated “from their native garb to an English dress” as to be often unrecognizable (Reference WilsonWilson 1855, ii). Yet Wilson does propose using a diacritically expanded Roman alphabet in order to enforce a standardized representational system for local scripts such that variants like luckeradge for lakhiraj (which he glosses as “without tax”) “should no longer be perpetrated or tolerated.” The goal of glossaries like those of Wilson and Wilkins, then, is not to domesticate Hindi and Urdu words, but rather to identify, transfigure, and regulate those terms that have “thickly studded” colonial documents in order to appropriate them in service of the East India Company’s profits and the larger colonial project.

14.3.2 Defining Colonial Subjects

The appropriating of local words to the needs of colonial rule is a goal that Arthur Burnell calls “purely technical” in his “Introductory Remarks” to Hobson-Jobson: Being a Glossary of Anglo-Indian Colloquial Words and Phrases, and of Kindred Terms (1886; 2013), which he authored with Henry Yule. Burnell writes otherwise admiringly of Wilson’s glossary in particular, but his characterization signals his belief that Hobson-Jobson is something altogether different than an interpretive aid to facilitate colonial administration. Instead, its stated purpose is to document not only those words that “recur constantly in the daily intercourse of the English in India, either as expressing ideas really not provided for by our mother-tongue, or supposed by the speakers (often quite erroneously) to express something not capable of just denotation by any English term,” but also words “that seem to have been admitted to full franchise” into English. Its intended scope, therefore, is a comprehensive catalog of borrowings – the history of the British in India as filtered through the lens of language contact.

In Yule and Burnell’s telling, those intertwined histories rationalize colonial authority. Much like The Ladies Dictionary and its encoding of patriarchal authority, Hobson-Jobson’s ideological positions are often advanced in its definitions. Ari Singh Reference AnandAnand (2011) points to entries like the one for Coorg: a territory “which was annexed to the British Government in consequence of cruel misgovernment in 1834” (Reference Yule and BurnellYule and Burnell 1886, 194). As Anand observes, the entry’s description of historical events is of dubious accuracy and serves to represent the British takeover as a natural and benevolent response to supposedly irresponsible self-rule. Other entries similarly work to implicate colonial subjects in their own subjugation. For example, there are roughly thirty-five entries describing and sorting servants into a kind of taxonomy. These include entries for headwords like bungy ‘house sweeper,’ dubash ‘interpreter,’grass-cutter ‘horse keeper,’ kitmutgar ‘meal server,’ kittysol-boy ‘umbrella carrier,’ and bheesty:

The definition imagines that “Indian servants rejoice” at being named a bheesty or water carrier for the British. While in and of itself this bit of psychological projection is a self-serving fiction, it is followed by descriptions of them as “diligent,” “faithful,” “unobtrusive,” and “uncomplaining.” Such adjectives were routinely attached to servant archetypes who defended British colonists from rebellious natives in newspaper stories, novels, and adolescent fiction, particularly after the uprising of 1857 and the emergence of the “Mutiny Novel” as a genre (Reference ChakravartyChakravarty 2005). The trope of the “faithful ayah” – a maid or governess who protects her youthful British charges from a host of threats – was especially common (Reference BrownBrown 2018, 182–183). The entry bheesty taps into these circulating representations not only with its choice of adjectives, but also in its closing anecdote conjuring a battlefield onto which would-be carriers venture to supply wounded British soldiers “in the face of much personal danger.”

14.3.3 Imagining Vocal Cultures and Inventing Speakers

Such imagery imbues Hobson-Jobson with a sense of colonial nostalgia, despite the fact that it was conceived in 1872 during the early part of the British Raj and published less than a decade after Queen Victoria was crowned Empress of India. It is a nostalgic stance that is amplified in dictionaries published in the early twentieth century (a time when Hobson-Jobson is revised and reissued). For example, A. P. Hill begins Broken China: A Vocabulary of Pidgin English (Reference Hill1920), which he published in Shanghai:

The focus of Hill’s dictionary is Chinese Pidgin English, a variety that emerged from the conditions of language contact in Guangzhou when the city was the center of trade between Europe and China in the eighteenth and nineteenth centuries. The dictionary contains headwords that are common in earlier glossaries like chow-chow, joss, and maskee, as well as others that reflect the dictionary’s less-than-strictly lexicographic interests like East India Company, fingernails, and missionaries. Similar to Hobson-Jobson, the definitional content of the entries in Broken China can express a mixture of charisma, paternalism, and anxiety. An underlying sinophobia is evident, for example, in the definition for Pidgin English, which begins with a description of its phonology and syntax, then goes on to equate the limitations of the language to the “paucity of transmissible thought” of its speakers:

Hill’s dictionary was preceded by another purported record of Chinese Pidgin English that was even more pointedly sinophobic: Pidgin English Sing-Song (Reference LelandLeland 1876). That work’s blunt racism is underscored in the concluding sentence of its preface, where the author, Charles Leland, like Hill draws a direct line between the language variety and the psychology of its speakers. But Leland goes even further in pathologizing and infantilizing not only Chinese, but also African diasporic vocal cultures in his assertion that Chinese Pidgin English “can present no difficulty to any one who can understand negro minstrelsy or baby talk.” Following the preface are poems and stories, which are followed, in turn, by a glossary. The former are composed in a literary dialect style popular during Gilded Age America, like this parody of Little Jack Horner:

Most of the poems and stories appear to have been written by Leland himself, though they are passed off as authentic examples of the variety. Many of these ventriloquized artifacts are recycled in A Dictionary of Slang, Jargon & Cant (1889), which Leland created with Albert Barrère. In that dictionary, entries for headwords labeled “pidgin-English” are almost universally exemplified with uncredited excerpts from Pidgin English Sing-Song, as in the entry for chow-chow:

The contrived and exaggerated nature of the texts advertises a thinly veiled mockery of the variety and its speakers. Such mockery stands in contrast with the technocratic functionalism of Wilkins and Wilson. The difference is at least partially attributable to the fact that Leland was not a colonial bureaucrat, but an American who had never visited the region where the variety was spoken. That, as Kingsley Reference BoltonBolton (2003, 188) observes, did not dissuade Leland from proclaiming his preeminent expertise. In a letter to his niece, Elizabeth Robins Pennell, following the Ninth International Congress of Orientalists in 1891, he boasts:

Even if we cast a skeptical eye on Leland’s claims, Pidgin English Sing-Song was popular with the domestic audience for which it was intended. In this way, not only did out-group imaginings of the variety come to stand in for in-group use, but also the associated ideologies of difference and deficit were reinforced. By implication, those ideologies sanctioned the superiority of a perceived “standard” and the authority of its speakers. In late-nineteenth-century America, such authority was being marshaled against Chinese immigrants, and Leland’s works were influential in catalyzing attitudes and ultimately abetting the passage of the Chinese Exclusion Act of 1882 (Reference BoltonBolton 2003, 188–189). The national paranoia that fueled the act’s passage is explicitly articulated in a letter Leland wrote to Mary Alice Owen, a fellow folklorist and literary dialect enthusiast, in 1899:

Leland’s glossary, therefore, became propaganda for legal and legislative action – in this case for the exclusion and disenfranchisement of Chinese immigrants and communities. Thus, it neither exerts colonial control nor advances a nostalgic vision of the social order that such control enforced. Instead, it articulates the racial and national anxieties unleashed by the anticipated collapse of that order.

14.4.1 “Good” Words

The relationship between Leland’s glossary and legal authority was ancillary, contingent on the ideological resonances between his sociolinguistic anxieties and a growing sinophobia in the anglophone world. Other dictionaries, however, have been empowered more directly, perhaps none more so than “Webster’s” in its various editions. Before looking at “Webster’s” role in the American judicial system and the beliefs that have sustained that role for two centuries, it is useful to briefly outline its contents and its place in the American cultural imagination. For this purpose, we will make use of two editions. The first is the 1844 revised edition (Reference WebsterWebster 1844), which has been digitized by the Emily Dickinson Lexicon (Reference HallenHallen 2009). The second is the 1913 edition of Webster’s Revised Unabridged Dictionary (Reference Porter1913), which has been digitized both by the GNU Dictionary Project (Reference DyckDyck 2015) and Project Gutenberg. In broad strokes, they demonstrate “Webster’s” indebtedness to lexicographic traditions. Even the later and larger Reference Porter1913 edition bears similarities to the first hard-word dictionaries in emphasizing the rare and technical over the everyday and spoken. Both also hew closely to Samuel Johnson’s template in their citational patterns, contributing to the canonization of a select subset of authors. Moreover, some of these alignments suggest Noah Webster’s specific imaginings of American English and its relationship to a “pure” linguistic past.

That these patterns persist into the twentieth century – seventy-five years after Webster’s death – speaks to Webster’s enduring influence, so we will begin there: with a brief inventory of the 1913 edition. Although much larger and more general than early hard-word dictionaries, Webster’s Unabridged exhibits an analogous preference for uncommon, specialized, and technical words. One way to underscore this preference is to measure its overlap with language-in-use using data from the Corpus of Historical American English (Reference DaviesDavies 2010–). Of the dictionary’s more than one-hundred thousand unique headwords (not accounting for sense disambiguation), only 52 percent appear in the COHA data between 1810 and 1909 – the century before the dictionary’s publication. That number rises to 62 percent if all of the corpus data are included in the search (1810–2009). To be sure, most words are infrequent and, at just over four-hundred thousand words, COHA is not particularly large. Nonetheless, that leaves roughly forty-thousand orphaned headwords, many of which appear to be Latinate neologisms that seemingly have their origins in other dictionaries – occurring rarely, if ever, in other kinds of texts. Take, for example, abacinate ‘To blind by a red-hot metal plate held before the eyes.’ The definition in Webster’s is almost identical to the one in Murray’s New English Dictionary (Reference Murray and Murray1888), and it similarly appears in The Century Dictionary (Reference WhitneyWhitney 1889), The Progressive Supplemental Dictionary of the English Language (Reference FallowsFallows 1885), The American Agriculturist Family Cyclopaedia (Reference JuddJudd 1885), Chambers’ Condensed Encyclopedia (Reference StuartStuart 1895), The Imperial Dictionary (Reference Ogilvie and AnnandaleOgilvie and Annandale 1882), and The National Standard Encyclopedia (Reference BurtBurt 1888). Yet there is little evidence of its use in English outside of these and similar reference works.

In addition to its presences, Webster’s Unabridged possesses similar absences. For one, it largely eschews taboo words, labeling only eleven words like piss and dickens “vulgar.” Moreover, it omits euphemisms like gosh and golly that, as the COHA data show, had been circulating in American English for decades. These omissions certainly reflect what Edward Reference Gates, Hyldgaard-Jensen and ZetterstenGates (1990) terms “Webster’s” preference for “good” as opposed to “bad” words. However, they also seem to illustrate a disfavoring of words more associated with speech or fictional representations of speech. It is an imbalance hinted at in an accounting of its labels. While there are slightly more than 800 entries labeled “colloquial,” “local,” “slang,” or “vulgar,” there are more than 18,000 labeled “obsolete,” “rare,” or “archaic.” Not that the 1913 edition ignores speech-related words altogether. It includes, for example, entries for orthographic representations of regional pronunciations, such as worrit for worry and a glossed as a variant “of have, of he, and sometimes of it and of they.” Attending the a entry is an illustrative quotation from Shakespeare: “So would I a done.” Alternatively, worrit has no accompanying citation, but it is a headword that circulates in a number of nineteenth-century dictionaries like the Imperial and the Century with illustrative quotations from Charles Dickens, who is cited in Webster’s Unabridged more than 200 times. Such provenance is likely the reason that these entries are included, though their non-standardness is still clearly marked – with a ‘have’ termed a “barbarous corruption” and worrit labeled (like ain’t) “illiterate.”

Lacking that kind of provenance, other words like yep, nope, wow, and feller have senses that are not included, though they are circulating in American English during the nineteenth and early twentieth centuries.Footnote ³ The stark difference between the ascendant trajectory of feller and near-zero frequency of worrit, for example, can be illustrated using data from Google Books (see Figure 14.2). As one might suspect from its trend, feller was widely used in a variety of genres including fiction in the decades prior to the publication of the 1913 edition of the dictionary. In eliding contemporaneously and locally circulating variants in favor of others distanced by time and place, Webster’s Unabridged looks backward and across the Atlantic for many of its exemplars of speech. In some sense, this might seem at odds with Noah Webster’s interest in defining and shaping an “American” English as distinct from a “British” one, through both his lexicography and his schoolbooks (Reference RollinsRollins 1976). However, Webster had an Edenic view of American speech – conceiving it as closer to a Shakespearean ideal than British English, which had, in his view, become corrupted over the course of the intervening centuries. “If spoken American sounded like written Shakespeare,” Richard Bailey (reflecting on Webster’s endorsement) observes, “all the better for America” (Reference Bailey, Bergs and Brinton2017, 14).

Figure 14.2 Scatterplot showing frequencies (per million words) of feller and worrit in American English 1800–1909, using data from Google Books.

14.4.2 “Good” Books

That feller does not appear as a variant of fellow in the 1913 edition is perhaps even more surprising given that it was frequently used by popular American authors like Mark Twain. Twain, in fact, is mentioned only once in the edition and is not quoted at all. The authors that do get cited in both the 1913 and the 1844 editions closely mirror those of a dictionary published a century-and-a-half earlier: Samuel Johnson’s watershed A Dictionary of the English Language (1755). A comparison of the most commonly referenced sources in both editions of Webster’s is presented in Table 14.1, with the counts of Johnson’s citations (excepting those from the Bible) quoted from Reference 756WillinskyWillinsky (1994, 212). The biblical citations were estimated from an OCR version of Johnson’s dictionary.Footnote ⁴

In their similarities, the citational patterns illustrate the powerful ways that dictionaries have contributed to the construction of an English language literary canon that is almost exclusively white and male. (The two most frequently referenced women in Webster’s Unbridged, George Eliot and Elizabeth Barret Browning, are ranked ninety-first and ninety-seventh respectively.) For Johnson, the authorial taxonomizing was largely the point, as he saw his project as curatorial, devoted to preserving and promoting exemplary practitioners of style. Webster’s choice of citations served to advance his linking of American English to a “purer” linguistic past. Just as the words that are included (or excluded) reflect Webster’s idealization of American English, so too do the authors the dictionary cites (Reference Bailey, Bergs and BrintonBailey 2017, 13–16).

The citational patterns further reveal that, for Webster, American linguistic exceptionalism had a moral dimension. His evangelical cause is evident in a relatively greater emphasis on the Bible (Reference Hallen and SpackmanHallen and Spackman 2010). It manifests, too, in the content of his definitions. Entries for headwords like religion, nature, and backslider contain direct references to God and faith (Reference FineganFinegan 2020a). Less directly, the citational force of the Bible also affects definitional content. Consider, again, feller. While there is no sense given for a pronunciation of fellow, both the 1844 and 1913 versions do include feller, defining it as “one who hews or knocks down” and citing Isaiah. The persistence of that biblically sourced definition into the twentieth century is one small instantiation of the larger pattern illustrated in Table 14.1. Not only does it underline “the linked centrality of faith and language” for Webster (Reference FineganFinegan 2020a, 160), it also suggests the durability of that linkage – the success of Webster’s project to embed that linkage into the national identity. Moreover, in American culture, the notional linking of Christian religious authority has superficially implicated “Webster’s” (sometimes generically, occasionally to specific editions) as an arbitrator and guide. We heard echoes of this in the quoted review of the Third from this chapter’s introduction, which equates “the dictionary” to the Bible. That association was hardly unique, according to Edward Reference Gates, Hyldgaard-Jensen and ZetterstenGates (1990), as it was common during westward expansion for Americans to bring only two books with them: the Bible and “Webster’s.” Gates’ assertion is attested to by at least one woman’s memory of her family’s journey: “Our library consisted of the Bible and Webster’s dictionary – the only two books we felt we could not do without, and to make sure of them we brought them all the way across the plains” (Reference JudsonJudson 1925, 95).

14.4.3 “Webster’s” in the American Courts

That kind of perceived indispensableness was, of course, good for business, and the company that published “Webster’s” did not shy away from asserting its essential place not just in the American home, but also among American institutions. An advertisement that circulated in 1910 proclaims: “THE reference work, which has been tested, approved, and accepted by leaders of the world’s thought, action, and culture, – The Standard for the Schools, the Courts, and the Press” (A New Creation). The ad may seem a piece of marketing hyperbole, yet “Webster’s” has served important functions in American institutions, none more so than the courts (Reference WeisWeis 1987). Indeed, shortly after its initial publication in 1828, one can find examples of its being used in judicial decisions to arbitrate the meaning of key words, as in an 1831 case (Ware v. Gay) in which a man was injured when a wheel came off a stagecoach he had hired. The Supreme Court of Massachusetts decided in favor of the injured party, with the case partly hinging on the meaning of the verb to provide:

Other cases have looked to “Webster’s” for definitions of onion, explosion, concubine, and chiropractic. The US Supreme Court case of Nix v. Hedden in 1893 famously rested on the definitions of fruit and vegetable and which category tomato belonged to. In Nix, the court actually rejected the dictionary definitions, which seemed to suggest tomatoes are fruit. However, that kind of skepticism has been more the exception than the rule. As in Ware v. Gay, the dictionary generally conferred authority over meaning, lending it extraordinary power in the determination of statutory intent.

The trend has been toward affording dictionaries more authority not less, as legal theories like “textualism” and “originalism” have come into vogue (Sonpal 2003). To test the rate at which “Webster’s” has been referenced over time, Figure 14.3 combines data from the Federal Judicial Center (2018) and LexisNexis, showing the number of terminated federal cases in which the words Webster’s and dictionary are both present. The results track with accounts of the US Supreme Court’s increasing reliance on dictionaries during the Rehnquist Court in the 1990s (Reference Thumma and KirchmeierThumma and Kirchmeier 1999) – a trend that continued into the twenty-first century. Between 2000 and 2010, the US Supreme Court used dictionaries to define 295 words, and among general usage dictionaries Webster’s Second and Webster’s Third were particularly popular, being cited in forty and sixty opinions respectively (Reference Thumma and KirchmeierThumma and Kirchmeier 2010). By Thumma and Kirchmeier’s accounting, various editions of “Webster’s” (which also include An American Dictionary of the English Language, Webster’s Collegiate Dictionary, and Webster’s International Dictionary) were cited in 137 cases to define 154 unique words, including arms, capacity, cruel, force, if, until, and would.

Figure 14.3 Rate per decade at which terminated US federal court cases include Webster’s and dictionary in decisions.

The accelerating judicial reliance on dictionaries highlights beliefs about linguistic meaning, about its location within texts, and about the ability of dictionaries to aid in its recovery, to impartially referee disagreements and to remove ambiguity. As one critic of the Supreme Court’s use of dictionaries points out, their appeal is obvious, as “dictionaries present an aura of objective authority” (Reference RubinRubin 2010, 206), and the general appeal of “Webster’s” attests to the institutional and cultural authority it has cultivated from its beginnings. The ideological import of these practices, however, extends beyond the generic authority conferred on dictionaries or even the specific preference that American courts seem to have for “Webster’s.” The authors who studied the citational practices of the US Supreme Court argue that the specific edition of “Webster’s” that a given judge reaches for is weighted with ideological meaning. They note that the most frequent citer of dictionaries, Antonin Scalia, had a stated preference for Webster’s Second because of the Third’s “permissiveness” (Reference Thumma and KirchmeierThumma and Kirchmeier 2010, 96–97). He cited the Second in twelve cases between 2000 and 2010, while citing the Third in only a handful in the 1980s and once in the 1990s, indicating how his view of dictionaries aligned with a more prescriptively oriented edition than with the wayward “guardian” of the Third.

While the relative appeal of those mid-twentieth century editions is an artifact of debates over whether dictionaries should prescribe or describe usage and while the generic appeal of “Webster’s” is related to its historically promoted authority, the appeal of another commonly cited edition appears to be amplified by its role in shaping America’s linguistic and national identity. That dictionary, the 1828 edition, is often cited by the Supreme Court “as evidence of the original meaning of the Constitution” (Reference MaggsMaggs 2013, 389). It is, in other words, a favorite among judges who follow an “originalist” philosophy, which purports to extract “objective” meaning from historical documents like the United States Constitution. In the context of that enterprise, dictionaries are understood as lenses through which one can read past texts and impartially determine meaning. Scalia, a leading originalist, referenced the 1828 edition (or its various reprints) in seven decisions between 2000 and 2010 – defining thirteen words, such as affidavits, contrivance, keep, press, and procure.

How this conceptually positions the 1828 edition is perhaps clarified by comparing its role in the US courts with its place in another area of contemporary American life where it has gained traction: the Christian homeschooling movement in the US (Reference ButlerButler 2013). In that movement, the popularity of the edition is clearly related to its numerous biblical references and to the religious beliefs of its compiler – a connection that is not made, at least as explicitly, by originalists. It is, however, also related to the notion that the dictionary connects its readers to a “citizen” identity located in America’s past. According to Jeffrey Ian Reference ButlerButler (2013, 88), reading founding-era texts like Webster’s “kind of infused that sense of citizenship that was there in the schoolrooms at that time.” The use of the 1828 edition among homeschoolers and its use among originalists on the Supreme Court, therefore, share a sense of an idealized civic history that can be accessed either through the dictionary directly or via the dictionary as a conduit. This image of the “dictionary” and its relationship to language and culture provides an interesting counterpoint to the one articulated by J.S. in the preface to Glossographia. In that context, it was imagined as serving as a bulwark against a rising tide of perceived linguistic and social chaos. In this one, it is imagined not so much as a way to arrest change, but as a way to circumvent it and recover the past and a romanticized legal, civil, political, and cultural order. (For more about Webster and the role of his 1828 dictionary in Christian education, see Finegan, Chapter 19, this volume.)

15.2.1 There Are Cultural Prerequisites to Dictionary Making and Use

Dictionaries typically come into being in cultures that write, that have established systems of trade and education, and that need to translate between multiple languages. The oldest dictionaries of English are Latin–English glossaries from the early Middle Ages (450–1100 ce), which helped English speakers learn Latin and participate in the literate life brought to England by Christian missionaries from Rome. Geoffrey the Grammarian’s 1440 Promptorium Parvulorum is one such English–Latin vocabulary, a children’s storeroom, as its title has it, of some 10,000 words.

It wasn’t until the seventeenth century that the first monolingual dictionaries of English appeared. Because only cultures that value their own language crave or create dictionaries describing it. Long the colonized other of Latin and (Norman) French, later the ugly duckling to Romance languages with more formidable literary traditions, English wasn’t sufficiently respected by its own speakers until the early modern era, at which point schoolmasters like Richard Mulcaster started to suggest that it would be “a thing very praiseworthy […] and no less profitable than praiseworthy, if some one well learned and as laborious a man, would gather all the words which we use in our English tongue […] into one dictionary” (Reference MulcasterMulcaster 1582, 166). A few decades later, monolingual English dictionaries appeared. Robert Cawdrey’s Table Alphabeticall (1604) was among them; its preface implores Englishmen to speak, not “French English” or “English Italianated,” but rather “their mothers’ language,” the “King’s English” (Reference CawdreyCawdrey 1604, A3r–A3v). There are, then, some cultural prerequisites to dictionary creation and circulation, and different kinds of dictionaries have different prerequisites.

15.2.2 There Are Cultures Disinterested in Dictionary Making and Use

Some cultures are largely disinterested in dictionaries, and this is likely because dictionaries are hospitable and appropriate only for some languages and some social groups. Disinterest in dictionaries is not infrequently tied to experiences of dictionaries as damaging.

Speakers of entirely or predominantly oral languages, for example, may have little interest in lexicography because dictionaries, particularly pre-internet dictionaries, are artifacts of literate cultures: “Headwords, definitions, and various lexicographic notations are written; users are readers; entries are sequenced alphabetically; a major source of evidence for lexicographic decisions, especially in the European-American tradition, is citation from printed sources” (Reference Frawley, Hill and MunroFrawley et al. 2002, 8). Many languages have no written tradition until dictionaries themselves are compiled. The Navajo language’s earliest written documents, for example, were nineteenth-century wordlists by Anglo-American military officers and Christian missionaries who had also devised the alphabets used (Reference EnochsEnochs 2006).

Occasionally disinterest in lexicography has to do with maintaining a secret language or an antilanguage, one used by speakers marginalized by and resistant to the society in which they exist. In these cases, dictionaries often constitute linguistic voyeurism that at once compromises the safety of those observed while inviting discriminatory mindsets among those who observe (i.e. dictionary makers and users). What results is, again, a kind of cultural hostility to lexicography that may be spurred by the cultural hostilities exerted by or through lexicography.

In early modern England, canting dictionaries were one such mechanism of linguistic voyeurism, granting members of the upper class a glimpse of poor and criminal classes assumed to be predatory. The unsigned Manifest Detection of the Most Vile and Detestable Use of Diceplay (1552) is an early instance, cataloguing “the body of [the] art” of dice gambling and its “peculiar terms” so that “young gentlemen [and] others suddenly enabled by worldly abundance” might use it to “avoid the danger” of being cheated while also recognizing “(as it were in a glass [mirror])” the “miserable ends that a sort of handsome gentleman” might come to if he were to take to diceplay (Reference AnonAnon 1552, title, A1v, and B3v). Defining and publicizing the terms, tools, and techniques of would-be rogues made it harder for them both to find one another and to dupe the dictionary reader. Cant speakers were not themselves keen on dictionaries. John Awdeley’s Fraternity of Vagabonds (1565) attests to this, an informant agreeing to name the qualities and craft of “the ruffling and beggarly” only anonymously because “if my fellows do know […] That thus I did, they would kill me” (Awdeley 1575, A1v). The film Ball of Fire (1941) suggests an ongoing aversion to lexicography by working and criminal classes. When a lexicographer asks mobster moll and nightclub performer Sugarpuss O’Shea, “Would you object if I used you for observation in my study?” she replies, “Shove in your clutch.” Like other reluctant participants in the study, O’Shea is wary of anybody who wants to “move in on my brain” and fairly confident that a dictionary works something like a reform school, a confessional, a prison, or some combination of the three.

The relative sparsity of dictionaries of African American English is likewise tied to disinterest and suspicion. Though the history of this language stretches back to the seventeenth- and eighteenth-century relocation and enslavement of African people to the US, there were no dictionaries of Black Englishes for many decades. Not only is this rooted in legal prohibitions against Black literacy, scholarship, and authorship, it’s also symptomatic of a fundamental function of the African American “lexicon […] developed by giving special meanings to regular English words, a practice that goes back to enslavement and the need for a system of communication that only those in the enslaved community could understand” (Reference SmithermanSmitherman 1994, 5). One of the earliest lexicons of Black English, one focused on jazz jargon, was Cab Calloway’s Hepster’s Dictionary (1938) which ran to multiple variously titled editions within seven years’ time, complemented by quiz books and advertised in several of Calloway’s songs, one of which declares:

Calloway may have been keen to “spread the gospel of jive” in dictionary form (quoted in Reference ShiptonShipton 2010, 127), but subsequent dictionaries of African American Englishes – general or specialized – have been relatively scarce, particularly those penned by Black authors. Clarence Major’s Dictionary of Afro-American Slang (Reference 684Major1970), Heremese E. Robert’s The Third Ear: A Black Glossary (Reference Roberts1971), and Geneva Smitherman’s Black Talk (Reference Smitherman1994) are notable exceptions, all of which suggest that their dictionaries are documenting an “in-group, closed communication system” of Black Americans that is sufficiently “absorbed” into Eurocentric American English as to no longer pose a threat to community shibboleths (Reference 684MajorMajor 1970, 11; Reference RobertsRoberts 1971, unnumbered “Introduction”; Reference SmithermanSmitherman 1994, 5 and 17).

These examples signal that dictionaries hold a more tenuous place within some cultures than others, often for good reason. Dictionaries are always tied up in power differentials and divergent ideologies – ideologies not just about the dictionary and “the predominant roles dictionaries play in society” but also ideologies about “language, authority, literacy, education, commerce, and the interactions of these and related areas” (Reference MurphyMurphy 2018, 2). Groups denied authority, literacy, education, autonomy, and humanity are not likely to see dictionaries as a means of securing these things.

15.2.3 There Are Cultural Prohibitions to Dictionary Making and Use

Even in cultures that are largely enthusiastic about – rather than skeptical of – dictionaries, one finds prohibitions of dictionary making. Take, for instance, that early modern moment of lexicomania. Those who thrived in the “age of dictionaries” knew others who were excluded from it. When Samuel Johnson used this very phrase, the year before publication of his own dictionary, he was writing to a printer about a philosophical dictionary compiled by Johnson’s longtime companion Anna Williams: Williams and Johnson wanted to know if the printer thought “her dictionary likely to shift for itself in this age of dictionaries”; the answer was apparently no because no edition appears to have been printed (quoted in Reference LarsenLarsen 1985, 29). In 1771, Anne Fisher’s An Accurate New Spelling Dictionary and Expositor of the English Language, a formidable addition to her already lucrative educational reference empire, likewise ran into difficulty shifting for itself. An aggrieved former editor incited a rival dictionary maker and his publishers to file for an injunction on false grounds of plagiarism attested in a letter forged in Fisher’s hand; publication was stayed by two years, during which time, as Fisher predicted, the plaintiffs plagiarized her suppressed dictionary in order to “vamp up” their own (Reference Rodriguez-Álvarez and María EstherRodríguez-Alvarez and Rodríguez-Gil 2006).

Such instances show that the culture of dictionary making can sometimes be cordoned in accordance with social ideologies. In this instance, pervasive notions of sexism set femininity at odds with authority and thereby discouraged, and in some cases successfully prevented, women from pursuing publication of dictionaries. These a priori prohibitions have had lingering effects on the genre, its privileged practitioners cultivating and perpetuating practices that align with their own attitudes, values, habits, and goals.

15.3.1 Cultural Information and Cultural Perspectives May Be Lexicographical Assets

The kinds of dictionaries that wind up on brightly colored tables of giftable books tend to exude culture and character. Their titles alone promise readers an inside view of any number of distinctive social worlds: Hash House Lingo: The Slang of Soda Jerks, Short-Order Cooks, Bartenders, Waitresses, Carhops and Other Denizens of Yesterday’s Roadside (Reference Smiley and DicksonSmiley 1941/2012), The Language of World War II (Reference TaylorTaylor 1948), The Marriage Dictionary (Reference CareyCarey 1993), Slayer Slang: A Buffy the Vampire Slayer Lexicon (Reference AdamsAdams 2003), The Film Snob*s Dictionary (Reference Kamp and LeviKamp and Levi 2006), A Dictionary of Bullshit: A Lexicon of Corporate- and Office-Speak (Law 2006), The K★Pop Dictionary (Reference KangKang 2016), The Gen Z Dictionary (Reference KingKing 2020), The Cocktail Dictionary (Reference JeffreysJeffreys 2020), The Queen’s English: Dictionary for LGBTQIA+ Lingo and Colloquial Phrases (Reference DavisDavis 2021). These kinds of dictionaries, often but not always written by subject insiders who are lexicographical outsiders, are innumerable, indexing widespread interest in dictionaries as a mechanism for honoring and advertising the specialized language that animates specialized circumstance. For these dictionaries, culture is content and selling point.

CB dictionaries, which capitalized on a citizens band radio boom, were a fad in 1970s America. Lanie Dills’s “Official” CB Slanguage Language Dictionary (Reference Dills1975) topped The New York Times bestseller list, beating out The Joy of Sex and Our Bodies, Our Selves (NYT 1976, 221). “CB’s fascinatingly colorful lingo” is snugly situated in the culture of professional long-haul trucking and heavily coded – smokey for ‘cop,’ big slab for ‘expressway,’ swindle sheet for ‘trucker’s log,’ chew ‘in’ choke for ‘restaurant,’ pregnant roller skate for ‘Volkswagen Beetle’ (Reference DillsDills 1975, 5; Official CB Dictionary 1976). With nationwide oil shortages and speed traps, the jargon of truckers suddenly became the jargon of a whole “nation on wheels,” and Dills (a.k.a., Sugar Britches of Guitar Town) compiled her dictionary so that readers wouldn’t be “laughed off the channels” (Reference DillsDills 1975; Official CB Dictionary 1976, 9). The words and meanings in these dictionaries evidence a world with clear protagonists (big riggers and breakers) and villains (smokeys, mama bears, and Feds) as well as a worldview with clear priorities (travel, humor, independence, sex, personality, and anonymity).

Not only cultural content but also cultural formations can be assets to dictionary makers, many of whom occupy privileged positions within social hierarchies. Patriarchy and its attendant formations of gender, hierarchy, and domesticity, for example, have been crucial to many of the most ambitious dictionary projects. Male dictionary makers have regularly conscripted wives, daughters, sisters, lovers, and friends into helping with their dictionary making for little to no remuneration or recognition. Meanwhile, sex and gender stereotypes have successfully secured and circumscribed women’s participations in lexicography for centuries. Women’s limited professional opportunities and artificially low wages, as well as their alignment with domains of language whimsically deemed feminine, have made them a mainstay in an endeavor chronically understaffed, underfunded, and unable to document the vast diversity of Englishes. The English Dialect Dictionary (Reference WrightWright 1898–1905) is one such project to benefit from women’s free to affordable labor. The editor’s wife acted as a co-editor; the staff was almost entirely female as were many of its consultants; a third of its volunteer readers were women; and it drew material from more than a hundred published and unpublished dialect dictionaries and commentaries by women (Reference RussellRussell 2018, 116).

Sometimes lexicographers represent personal perceptions as if they were urgent cultural problems. Take, for example, the following titles which sound a similar alarm: Endangered Words: A Collection of Rare Gems for Book Lovers (Reference HertnonHertnon 2009), The Word Museum: The Most Remarkable English Words Ever Forgotten (Reference KacirkKacirk 2000), Let’s Bring Back: The Lost Language Edition (Reference BlumeBlume 2013) – these dictionaries style themselves as solving a problem (vocabulary death) and readers as helping in the effort (ostensibly by buying dictionaries; see also Rennie, Chapter 7, this volume). But supposedly forgotten, lost, and endangered words are and long have been safely ensconced in the texts that let us know they existed in the first place, not to mention any number of dictionaries that collected them thereafter. The so-called problem is as much an invention of the dictionary maker as its dictionary-shaped solution.

Sometimes lexicographers invent social problems with dictionary-shaped solutions. Such was likely the case with early monolingual dictionaries that invoked women audiences in their titles and prefaces: Thomas Blount’s Glossographia (1656, A5v), for instance, was “chiefly intended for the more-knowing women, and less-knowing men.” Scholars have claimed these first monolingual dictionaries were responding to a lexical deficit peculiar to women, but there is little evidence to suggest that women actually needed, wanted, or ultimately used those early dictionaries. It is, however, very clear that women constituted something of an imaginable “market” that dictionary makers named as they sought to bring the genre of the dictionary to public prominence. The idea of women as a needful and open audience was more important than any actual circumstance of gender-based linguistic disparity (Reference 746RussellRussell 2014, 36). It is, then, more accurate to see some appeals to cultural need as in fact constructions of it that reflect personal positions rather than social circumstances.

15.3.2 Cultural Information and Cultural Perspectives Are Sometimes Lexicographical Liabilities

General dictionaries are less readily marked as cultural artifacts because, by design, the prevailing privileged forms of professional lexicography tend to invite a sense of culturelessness. General dictionaries are typically made by lexicographical insiders performing as linguistic omniscients: They turn “protean entities” – the “shifting patterns [of words] that change from year to year, from locality to locality and from idiolect to idiolect” (Reference Ayto and HartmannAyto 1983, 89, 95) – to durable, transferrable explanations of meaning that might work for any number of speakers in any number of circumstances. Historical titles use vaunted adjectives to gesture to cultural transcendence: An Universal Etymological English Dictionary (Reference BaileyBailey 1721), A Complete Dictionary of the English Language (Reference SheridanSheridan 1798), A Complete and Universal English Dictionary (Reference BarclayBarclay 1782). And, over time, an accreted dictionary style – a “quasi-scientific tone” aberrant, abstract, esoteric, formulaic, and deviant from almost all other English registers – has come to position dictionaries as somehow outside of ordinary language and culture (Reference Moon, Knowles and MalmkjaerMoon 1989, 69).

Historical general dictionaries may have flagrantly vented “political and social prejudices” – Samuel Johnson was famous for disparaging the Scots in his definition of oats, for example; British dictionaries for disparaging American coinages – but these “peccadilloes,” once thought to “lend a distinctive charm” to dictionaries (Reference AlgeoAlgeo 1990, 214–215), became lexicographical liabilities in a more scientific era that understood personal perspectives and cultural flavor to be irrelevant, even anathema, to the empiricism, systematicity, and sobriety necessary to the discovery and documentation of universal knowledge. “By the mid-nineteenth century, earlier English dictionaries began to be criticized for their excessive subjectivity and prescriptivism, lack of scientific rigor, absence of systematic investigation, and limited coverage of the lexicon” (Reference Ogilvie and SafranOgilvie 2020, 54). And so new forms of dictionary making took shape, ones that aimed for descriptivism over prescriptivism and favored historical evidence over contemporary insight.

The Oxford English Dictionary is often positioned as the archetype of this ilk, a comprehensive and unrelentingly systematic catalogue of “English vocabulary from the time of the earliest records down to the present day, with all the relevant facts concerning their form, sense-history, pronunciation, and etymology,” embracing “not only the standard language of literature and conversation, whether current at the moment, or obsolete, or archaic, but also the main technical vocabulary, and a large measure of dialectal usage and slang” based on “several millions of excerpts from literature of every period amassed by an army of readers and the editorial staff” (Murray et al. 1989, vii). The OED is indeed ambitious in its aim, archive, army, and achievement. But even a net of the finest weave and widest compass will catch not just words but worldviews.

15.3.3 Cultural Information and Cultural Perspectives Are Always Legible in Dictionaries

Lexicographers take a range of stances as to the “proper” place of culture in the work of lexicography. One might say there is a methodological continuum in professional dictionary making that runs from what Prezemysław Reference ŁozowskiŁozowski (2018, 173) calls “culture-hostile” to “culture-friendly.” The forms of dictionary making most “hostile” to culture conceive of dictionaries as documenting language as a system: words are more and less closely related to other words, and it is the lexicographer’s job to scientifically, systematically, and objectively define words strictly by describing their central characteristics in contrast to the central characteristics of other relevant words. “Culture-friendly” lexicography, on the other hand, sees dictionaries as documenting language as symbolic of speaker experiences: words are signs, and it is the lexicographer’s job to define words by presenting the world of subjective, social, and cultural associations, attitudes, and experiences they evoke; in conjuring worlds, culture-friendly dictionaries often extend from word books to all-purpose references, complementing A-to-Z entries with maps and timelines; tables of weights, measures, and monies; portraits of literary luminaries or political leaders; and so on.

Of no small importance here is the fact that even the most culture-hostile dictionaries will contain cultural information as well as cultural bias. The OED is a notable example, intended and “generally regarded as an impersonal, objective document of comprehensive authority”; it stands as “an overwhelming symbol of scientific and objective scholarship,” a paragon of “modern lexicography” that systematically describes the lexicon on the basis of historical evidence (Reference BrewerBrewer 2005, 261; Reference Ogilvie and SafranOgilvie 2020, 54). Its pages nevertheless belie not only an imperial and ethnocentric sensibility but also “differing views” of the English language emerging from the “different personal histories and lexical environments” of different editors working in different times (Reference 702BensonBenson 2001; Reference BrewerBrewer 2005, 264). The idiosyncratic style of Robert W. Burchfield, editor of the Supplement to the Oxford English Dictionary (1972–86) has come under particular scrutiny, for instance. He penned “avowedly personal” prefaces, applied “usage labels and comments which tended towards subjective rather than objective judgement,” selected particularly pejorative quotations to describe female homosexuality, and departed in practice from inclusive policies on loanwords and World Englishes (Reference BrewerBrewer 2005, 270; Reference BrewerBrewer 2009, 275–277; Reference OgilvieOgilvie 2013, 21).

Cultural information and cultural biases are strewn throughout dictionaries, legible in entries (definitions, pronunciation guides, labels and notes, illustrative examples) as well as the paratexts that surround and extend A-to-Z content (prefaces, inserts, appendices, but also advertisements, press releases, websites, podcasts, social media feeds). Not all parts of dictionaries are equally or identically hospitable to cultural material, but they all play a part in recording and promoting cultural facts and affinities.

15.4.1 Dictionaries Can Legitimate Languages

“With a dictionary, a language (or language variety) is no longer a dependent or derivative, no longer insufficient and inadequate” (Reference BaileyBailey 2009, 279). In dictionary scholarship, one of the most widely touted roles of dictionaries is to assert linguistic legitimacy. By describing languages, dictionaries declare a language variety and its users adequate, sufficient, original, independent. For this reason, tools that assert linguistic legitimacy have been prominent also in asserting national, regional, ethnic, and social identities.

Many of early modern Europe’s emerging nation states formed language academies tasked with publishing dictionaries that would directly and indirectly declare the sovereignty of their empires by proclaiming the sufficiency of their vernaculars in contrast to the classical learned languages of Latin and Greek. Italy’s Vocabolario degli Accademici della Crusca (1611), France’s Dictionnarie de l’Académie française (1695), and Spain’s Diccionario de la lengua castellana (Reference Española1726–39) are examples of academy-authored dictionaries that aimed to “visibly and tangibly represent – with statuesque rigidity” – “the humanistic ideal of a timeless, perfect and absolute literary language, a model language” of a model nation (Reference NencioniNencioni 1990, 345, 349). English may not have had an academy, but it had Johnson’s Dictionary and, later, the Oxford English Dictionary, dedicated to Queen Victoria and hailed by King George V as “a unique monument to British learning and enterprise” (Guardian 1928, 6).

A great many English-language national dictionaries followed, each intended to be, and widely recognized as, a declaration of cultural if not political independence: John Jamieson’s Etymological Dictionary of the Scottish Language (Reference Jamieson1808) was meant “to preserve and illustrate the language and early literature of a brave people” who were faithful to the United Kingdom (evidenced by the dictionary’s dedication to future King, George IV) while still “patriotic […] in defense of national independence” (evidenced by the dictionary’s insistence on Scottish as a language in its own right) (Reference JamiesonJamieson 1808, dedication, iv). Noah Webster’s An American Dictionary of the English Language (Reference Webster1828) was spurred, at least in part, by his opinion that “the genius and character of my countrymen” is debased by “the implicit confidence they place in English authors, and their unhesitating submission to their opinions, their derision, and their frowns” (quoted in Reference BaileyBailey 2009, 282). “It is not only important, but, in a degree necessary,” Webster proclaimed, “that the people of this country, should have an American Dictionary of the English Language” (Reference WebsterWebster 1828, preface). The “necessity for a special book on Australasian English,” in the case of Edward E. Morris, and South African English, in the case of Charles Pettman, became apparent when they noticed the relative absence or inaccurate treatment of their English varieties in British and American dictionaries (Reference MorrisMorris 1898, x–xi; Reference PettmanPettman 1913, v). Morris’ Austral English: A Dictionary of Australian Words, Phrases and Usages (Reference Morris1898) and Pettman’s Africanderisms: A Glossary of South African Colloquial Words and Phrases and of Place and Other Names (Reference Pettman1913) were correctives that made national Englishes visible both locally and globally. When short an army and navy, a dictionary will apparently do just fine to make the difference between a dialect and a language.

Social collectivities other than nations have likewise turned to lexicography as a means of claiming humanity and sovereignty. Where national dictionaries tend to use them to reinforce borders and assert cultural coherence, regional and ethnic dictionaries often use them as occasion to grapple with blurred boundaries, internal divergence, and geographic dispersion. Take, for example, Librado Keno Vasquez and Maria Enriqueta Vasquez’s Regional Dictionary of Chicano Slang (Reference Vasquez and Vasquez1975). According to its preface, this “First Regional Dictionary of Chicano–Hispano American Slang (dialects) spoken in the Southwest (Diccionario Regional De Dialectos Chicanos–Hispanos Americano Habladas En El Suoeste) is the result of an urgent demand on the part of the tri-ethnic communities to have a rather more concise and extensive list of genuine regional speech variants that might be considered ‘essential’ in the education of our Spanish-speaking community” (Reference Vasquez and Vasquez1975, 5). Vasquez and Vasquez are careful to explain they’ve included what might appear to be “questionable” words that “no professional teacher would be willing to accept,” preferring to respect that the Spanish-speaker who deems certain words improper “in the education of their children” does “not represent the entire Chicano population” (Reference Vasquez and Vasquez1975, 5 and 7). Their dictionary’s documentation is therefore broadly inclusive precisely because it’s meant as recognition of and support to a multiethnic audience that is as ideologically diverse as it is linguistically diverse.

The linguistic legitimacy afforded by dictionaries can also further objectives of community building and spiritual reconciliation. Kalpulli Izkalli, a community group for detribalized indigenous New Mexicans, folded dictionary making and dictionary use into their project of “recovering our traditional language,” Nahuatl (Reference Corritore, Cruz, Gonzales, Hinojosa, Lugo, Lugo, Ledesma, Silva-Bañuelos, Rodríguez and ZamoraCorritore et al. 2000, 1). Their 2000 vocabulary and exercise book, Amoxizkalli: el Libro de Izkalli, was intended to “help us heal and be better human beings as we began to understand language and its connection to the land” (Reference Corritore, Cruz, Gonzales, Hinojosa, Lugo, Lugo, Ledesma, Silva-Bañuelos, Rodríguez and ZamoraCorritore et al. 2000, 1). Dictionaries can then – not just in their publication but in their very compilation – be a part of “community work for communal good” (Reference Corritore, Cruz, Gonzales, Hinojosa, Lugo, Lugo, Ledesma, Silva-Bañuelos, Rodríguez and ZamoraCorritore et al. 2000, 1; see also Reference Schreyer and TurinSchreyer and Turin 2023).

These examples show that dictionaries can indeed legitimate languages when dictionary makers are themselves keenly attuned to – often a part of – the communities whose language is documented.

15.4.2 Dictionaries Can Suppress Languages

While it’s fair to say that the vast majority of dictionaries are language positive – ultimately serving to honor the languages they describe and in some way empowering the speakers of those languages – it is important to note that this is not always the case.

Some colonial dictionaries were compiled, published, and used in such a way as to inaccurately represent indigenous languages, often exclusively for the use of colonizers, at the expense of the colonized, and sometimes in the service of language erasure. Throughout the nineteenth century, many Western European linguists perceived indigenous languages as inferior and resented having to learn or use them, let alone make dictionaries of them. In his Grammar and Dictionary of the Dakota Language (Reference Riggs1852), Presbyterian missionary Stephen Return Riggs writes, “The preparation of this volume […] was not premeditated but has been a result altogether incidental to our work. Our object was to preach the Gospel to the Dakotas in their own language, and to teach them to read and write the same, until their circumstances should be so changed as to enable them to learn English” (Reference RiggsRiggs 1852, v–vi). Riggs may not have enjoyed dictionary making but he enjoyed Dakota, noting that, after years of studying it alongside his wife Mary, it became “a heart-language”: “A Dakota word began to thrill as an English word” (Reference RiggsRiggs 1880, 106). However, official government policy was English only, so Riggs was forbidden to teach in the “barbarous dialect” of his heart and instead encouraged to look forward, as he does in his preface, to the day when English alone would fill the mouths of Indian school children (Reference 744ReyhnerReyhner 1993). Though he may not have intended it, his dictionary played a role in discouraging Dakota use; the language is now considered endangered.

Like cant dictionaries before them, many colonial dictionaries document language varieties deemed inferior and undesirable, and their projects of language documentation can tend toward ideological interpellation, suppressing and subjugating speakers more than celebrating and spreading their language.

15.4.3 Dictionaries Can Facilitate Social Mobility

Another oft-celebrated function of dictionaries in lexicographical scholarship is social mobility through autodidacticism; dictionaries are understood as “instruments for self-teaching” (Reference BéjointBéjoint 2000, 94).

The notion that the earliest English dictionaries were teaching aids is rooted in the fact that many bilingual dictionaries of the fifteenth and sixteenth centuries were commemorations of private education. For instance, Mary, eldest daughter of Henry VIII, had studied French from childhood with longtime court tutor, Giles DuWés; as an adult, she “right specially and straightly […] commanded and encharged [him] to reduce and to put by writing” the instruction he’d given both her and “her said progenitors and predecessors,” hoping the reference might “profit to others as to herself” (Reference DuWésDuWés 1533, A4r). A dictionary with dialogues, DuWés’ Introductorie for to Lerne to Rede, to Pronounce, and to Speake Frenche Trewly (1533), was the result: a tool of public education.

Monolingual English dictionaries followed suit. They’d been called for by schoolmasters (to schoolboys), but early English dictionaries addressed themselves to the marginalized masses, those who had limited access to formal or institutional education – language learners, tradespeople, women, children. These appeals appeared in titles and prefaces, as in John Kersey’s A New English Dictionary, Chiefly Designed for the Benefit of Young Scholars, Tradesmen, Artificers, and the Female Sex (Reference Kersey1702).

As Kersey’s title suggests, the notion that dictionaries are a mechanism of self-improvement and social mobility has been popular not just in dictionary histories but also dictionary marketing, and it would remain so over time. In twentieth-century United States, dictionary-ownership was often styled as a rung on the ladder to success. An advertisement for the Webster’s Seventh New Collegiate Dictionary (1965), a hefty tome handsomely dustjacketed in red, proclaimed “One of the differences between a white collar and blue collar could be a red jacket,” illustrated by a male model frowning with a shovel or smiling with a Webster’s (quoted in Reference MurphyMurphy 2018, 16).

Whether or not dictionaries are actually in any way capable of facilitating social mobility remains to be seen, let alone compellingly documented by research on dictionary use. Certainly, dictionaries did little to upset the structurally secured illiteracy of early modern women or the structurally secured economic insecurity of late-twentieth-century Americans.

15.4.4 Dictionaries Can Facilitate Social Discrimination

Competing with the idea that dictionaries solve social problems are instances in which they have been accused of creating such problems, particularly by perpetuating damaging social stereotypes. In 1969, a Manchester merchant, Marcus Shloimovitz entered into a protracted legal battle with Clarendon Press, publisher of Oxford dictionaries including the OED, regarding its definitions of Jew, which Shloimovitz maintained were “insulting, abusive, and an incitement to racial prejudice” (Reference HickmanHickman 1969, 5). Though Shloimovitz provided affidavits from other English Jews fearful of racial violence, he failed to prove personal damage had been caused by the definitions, and the case was dismissed by a High Court judge in 1973 (Reference EderEder 1973, 20). The Clarendon Press stood by the argument that “it should not be understood that lexicographers approve of the usages they report” (Reference EderEder 1973, 20).

Around the same time, critiques of mainstream dictionaries for derogatory representations of women and femininity were fulminating in the US. Intending to create a Feminist English Dictionary, former college teacher Ruth Todasco and a group of Chicagoans, collected from published dictionaries definitions describing women – lady, girl, grandmother, slut – and found the collection actually served as a list of English Words and Phrases Reflecting Sexist Attitudes toward Women in Patriarchal Society, a fact attested by the various titles under which the group published its list in 1973. “It took the dictionary to make me aware of the demeaning nature of being categorized as a subspecies,” explained a reader (Reference FarrellFarrell 1974, 33), affirming the disjunct that could be felt between daily experiences of meaning and catalogs of meaning in mainstream lexicography. Artist Pat Reference CourtneyCourtney’s (1988) compilation of illustrations from mainstream dictionaries – images attending to entries for anything from camera to spinster, hammock to yoke – likewise showed the incidental and explicit sexism of dictionaries; Courtney discovered that any woman is a passive woman when illustrated in a dictionary. These same “subtle and often destructive attitudes” were of concern to Maxine S. Rose, who appealed to fellow college teachers to exercise greater and more frequent scrutiny of references that female students are expected to use “daily as a research tool” (Reference RoseRose 1979, 375). Dictionaries of the twentieth century did little to formally address such concerns; revision of gender-biased lexicographical practice was rarely a policy priority at mainstream dictionaries and more likely a pet project quietly pursued by individual women editors. Twenty-first-century lexicography seems to have taken the matter more firmly in hand. A 2016 social media frenzy over “rabid feminist” – given as a usage example for rabid – compelled Oxford Dictionaries to recognize “the unintentional impact that authoritative resources like the dictionary can have on historically marginalized groups,” and, in 2020, they revised a selection of sexist definitions and example sentences in addition to establishing new editorial standards and practices (Reference FloodFlood 2020).

In the 1990s the National Association for the Advancement of Colored People called for a national boycott of Merriam-Webster dictionaries, which defined nigger as “a black person – usually taken to be offensive.” The NAACP took its cue from two Michigan women, one of whom, Kathryn Williams, a curator of an African American history museum, initially took exception to the definition when it proved of little help in “trying to assure a young person that the word ‘nigger’ cannot be rightfully applied to blacks”; instead, she found a “definition [that] is simply not accurate” (Reference FletcherFletcher 1997, A16). The boycott received widespread news coverage and was eventually successful in encouraging Merriam-Webster to add a usage note. The NAACP appears to have been less successful in a related objective to assure “economic reciprocity” by asking Merriam to “share with the NAACP their records on procurement, employment, promotion and the makeup of their board of directors to determine if a culture within the company has made it difficult for them to recognize why this definition is unacceptable to millions of Americans” (New York Beacon 1998, 2). As Merriam-Webster lexicographer Kory Stamper put it some two decades after the boycott, “the landscape of lexicography is still overwhelmingly beige” (Reference StamperStamper 2017, 167).

This was not the first dictionary campaign waged by the NAACP; as early as 1944, they’d asked the US War Department to discontinue use of a pocket-sized English–French dictionary with “offensive use and definition of the word ‘n–r’” leading to the “humiliation of Negro soldiers”; a G.I. had brought attention to the issue after experiencing the dictionary to encourage “exceptionally friendly and nice” Belgian “people to insult us” (Pittsburgh Reference CourierCourier 1944, 4). And it wouldn’t be the last campaign waged against Merriam-Webster for racially insensitive defining practices. In 2020, a recent college graduate, Kennedy Mitchum, annoyed by dictionary-citing social media scolds, encouraged Merriam-Webster to revise its definition of racism to explain that racism is not just prejudice (as their definition made clear) but “prejudice combined with social and institutional power. It is a system of advantage based on skin color” (Reference HauserHauser 2020). While being responsive to targeted revisions, Merriam-Webster and other mainstream dictionaries have often resisted larger lexicographical accommodations, declining to omit terms or to employ variable entry structures that might treat highly controversial vocabulary differently. In 2020, Dictionary.com publicized its decision to vary structure by separating their entry for Black (as it refers to a person) from black (as it refers to everything else). They recognized they were “breaking with dictionary conventions to group together words that share the same origin” but did so as a sign of cultural “respect and recognition,” “reflect[ing] Dictionary.com’s point of view that language entries have consequences” (Dictionary.com 2020).

The persistence of periodic public dictionary protest signals that dictionary users appreciate that words and meanings – as well as the reference works that document words and meanings – have the power to “cause harm, alienation, outrage, insecurity and alarm”; it likewise highlights that users occasionally ask their dictionaries to mitigate those effects by “becom[ing] more therapeutic and socially aware” (Reference McArthurMcArthur 1986a, 214). When dictionary makers have difficulty accommodating these requests, it’s not because they don’t personally wish to “distance themselves from attitudes which permeate their language and society,” but because the tenets of professional mainstream lexicography hold that being therapeutic or socially aware “implies authorial intrusion and subjectivity” inappropriate to the work of objective language description (Reference Moon, Knowles and MalmkjaerMoon 1989, 68 and 88). Dictionaries therefore readily accommodate protests when the available “solutions accord with established praxis and are contained in scope” – editing a definition, omitting an illustrative phrase, adding a usage note (Reference RussellRussell 2021, 244–245). But the field has been skittish about cultivating a sophisticated response to the larger issue: Dictionaries affect public perceptions as well as the lived experiences of actual humans. That is, dictionaries have inadvertent impacts, and, until dictionary makers take those inadvertent impacts seriously, they will be unable “to fully, ethically, reflexively, and relationally theorize how dictionary making should proceed” (Reference RussellRussell 2021, 245). Dictionaries could yet figure out how to meaningfully “promote intellectual engagement with words” by presenting facts, complications, and commentary allowing for “perspectives that may be directly, openly critical of one another” (Reference AdamsAdams 2007, 10 and 14). And they might thus make a start at theorizing “emancipatory” or “socially accountable” lexicography (e.g. Reference ChenChen 2019).

A useful addition to Łozowski’s rubric of culture-hostile and culture-friendly might then be culture-conscientious, a term to be invoked when referring to dictionaries that, taking an intralinguistic tack or an extralinguistic one, concertedly work to understand and mitigate the cultural consequences their descriptions of language can and do have, particularly for vulnerable groups like those accorded slurs or (largely pejorative) overlexicalization.

16.2.1 Good Usage

The most basic sense of usage refers to judgments of correctness between two or more variants, with one of the variants termed “good usage.” An example from the American Heritage Dictionary (AHD) comes in the entry irregardless: “it is a form that many people mistakenly believe to be a correct usage in formal style but that in fact has no legitimate antecedents in either standard or nonstandard varieties.” This good usage sense has long been an important part of dictionaries: Samuel Johnson did not hesitate to label some senses low, improper, vulgar, and so on (Reference Allen, Ali Jazayery, Polome and WinterAllen 1978; Reference WildWild 2009). Good usage became increasingly foregrounded after the publication of Webster’s Third New International Dictionary, Unabridged (hereafter Webster’s Third) in 1961, when numerous critics complained that the dictionary had abrogated its responsibility to give judgments about good usage (Reference Sledd and EbbittSledd and Ebbitt 1962; Reference MortonMorton 1994). The American Heritage Dictionary was created to restore the dictionary’s role as arbiter of correctness (Reference FineganFinegan 2003, 219) and is best known for its usage panel of over 100 experts who opined on usage issues. AHD also added an explicit section to its microstructure labeled “usage notes,” like the usage note for quick quoted above. A few earlier dictionaries had devoted separate sections to usage issues: Joseph Reference WorcesterWorcester (1860) used the pointing-finger icon (☞) and Webster’s Third – the same dictionary so roundly criticized for neglecting usage – used a dash to demarcate usage advice in some entries, while Random House (Reference Stein1966) used –Usage. Worcester does not discuss such usage notes in its front matter, while Webster’s Third and Random House do. Otherwise, most dictionaries before AHD had mixed their usage advice among their definitions, not as a separate section within the entry.

16.2.2 Actual Usage

The more transparent sense of usage, deriving from use, is captured in one of the senses for usage from AHD: “The way in which words or phrases are actually used, spoken, or written in a speech community.” Actual usage in dictionaries is sometimes explicitly invoked, as when AHD notes “Citational evidence clearly indicates that healthy and healthful have shared the meaning ‘conducive to good health’ since at least the mid-16th century. Therefore both healthy and healthful are correct in these contexts: a healthy (healthful) climate; a healthful (healthy) diet.” Actual usage can be both an authority and a challenge to good usage. Lexicographers and grammarians alike have long understood that in questions of variation, actual usage is the ultimate arbiter. Joseph Priestley is often quoted to show this understanding among eighteenth-century grammarians, where he uses custom in the sense of actual usage: “It must be allowed, that the custom of speaking is the original and only just standard of any language” (Reference PriestleyPriestley 1768, ix). Elsewhere, he writes “the thing [variation] must remain undecided till all-governing custom shall declare in favour of the one or the other” (Reference PriestleyPriestley 1761, vii). Priestley is articulating a reliance on custom or usage that had been noted famously by Horace:

At least in principle, actual usage has long been recognized as an authority for establishing good usage.

But for those who value correctness distinctions as a way of showing care and facility with language, actual usage is a challenge to the very notion of good usage. Wilson Follett, creator of his own usage guide, memorably characterized actual usage as “an organized assumption that language good enough for anybody is good enough for everybody” and that usage is “what a swarm of folk say or write by reason of laziness, shiftlessness, or ignorance” (Reference FollettFollett 1960, 73). The actual usage vs. good usage distinction has come to characterize descriptive vs. prescriptive approaches to language, in which a descriptive approach is said to focus on how language is actually used, whereas a prescriptive approach focuses on how language should be used.Footnote ³ Somehow the term usage can capture both sides of this is / ought dichotomy.

Actual usage is ubiquitous in dictionaries, but it largely remains implicit. At least since Johnson (Reference ReddickReddick 1996, 25–54), dictionaries have always relied on actual usage for their definitions: citation files, reader programs, and corpora are indispensable. Dictionaries also use empirical evidence for their advice on pronunciation and even, to a degree, on spelling. While dictionary entries seldom document their decisions – for example, with a note giving a quantified distribution of spelling, pronunciation, or senses – they do typically include illustrative quotations from the empirical record. No one involved with creating a dictionary disputes that their dictionary is based on actual usage.

16.2.3 Contextual Usage

Between actual usage and good usage – is vs. ought – lies much conceptual space, and usage has also referred to positions between these two poles. Knowing a word, for example, extends beyond knowing its meanings. What are the social connotations that become associated with a word? Who uses the word? Is it associated with certain groups of speakers? Is it associated with certain contexts, like more or less formal contexts? One label that dictionaries have used for all this contextual information about words is, of course, usage. Here usage emphasizes how, not just how much, a word is used. We might call this “contextual usage,” and it is invoked whenever dictionaries add contextual commentary to their entries, as when AHD adds the label Slang to its entry heavy lifting or Mus[ic] to heavy metal.

Early dictionaries tried to provide at least some information of this sort. Reference OsseltonOsselton (2006) explains that the seventeenth-century dictionaries used symbols and marks to provide such contextual information. Johnson interposed actual evaluative terms like popular or vulgar among his definitions (Reference WildWild 2008), and dictionaries today use a variety of labels to give contextual information. As Reference LandauLandau (2001, 217) explains, “Usage may also take in the study of limitations on use, whether geographic, social, or temporal.” Various classifications have been promoted for such “usage labels,” “status labels,” or “restrictive labels”; Reference BarnbrookBarnbrook (2005, 190–191) notes eight categories: (1) currency (e.g. archaic); (2) frequency (e.g. rare); (3) region (e.g. British); (4) technical (e.g. medicine); (5) normative (e.g. nonstandard); (6) connotative (e.g. jocular); (7) diastratic (e.g. low); (8) integration (e.g. foreign). These labels have long been popular in dictionaries and include much information that goes beyond notions of correctness.

This contextual sense of usage is still primary in notes included in learner’s dictionaries, which (perhaps regrettably) are also called “usage notes.” In general dictionaries, usage notes nearly always treat good usage, but in learner’s dictionaries they give information about the larger, contextual sense of usage, namely how a word is used in particular circumstances. The circumstances range from collocations to pragmatic and cultural information. Reference CowieCowie (1999, 170), for example, describes “usage notes” for learner’s dictionaries as containing “explanations and examples clarifying differences – grammatical and stylistic as well as semantic – between words of roughly similar meaning” and gives as an example a note that distinguishes synonyms of the verb refuse by the nouns they collocate with:

Reference Xia, Xia, Zhang and NesiXia et al. (2016) apply usage notes additionally to cultural notions that adhere to a word, like the different evaluation of individualism in Chinese and American cultures. A learner’s dictionary intended for Chinese-speaking learners of English may point out that unlike the strongly negative connotations in Chinese culture, individualism will have mostly positive connotations in American culture (Reference Xia, Xia, Zhang and NesiXia et al. 2016, 431–432). Their suggested note does not claim that the positive evaluation of individualism is somehow correct or incorrect; instead, it and others like it give information about contextual factors that attach to the word.

It is easy to see how this contextual notion of usage could shade into the correctness notions in good usage. In the first place, many of the contextual considerations are precisely those that have been the concern of good usage. If a sense is flagged because it is used by “low” speakers or used in conversation, instead of, say, educated writing, it has essentially been criticized for the same reasons that bad usage is criticized. More generally, the contextual labels used in general dictionaries nearly always mark words that somehow deviate from Standard English – that variety of English used for school, college, or other educated purposes. Words and senses characteristic of Standard English, the default variety in dictionaries, remain unmarked. Dictionaries do not waste space labeling words as “standard” or “typical of writing” or “educated”; they mark only the words that depart from this default (Reference BrewerBrewer 2016a). Even when usage labels don’t necessarily imply nonstandard uses, they are nonetheless marked, and readers may readily infer that the marked forms are somehow inferior. What might be meant as strictly contextual usage labels become monitory labels.

16.2.4 Traditional Usage

A fourth sense of usage is much less recognized but may be just as important as the other senses, and that is “traditional usage.” The crucial notion of traditional usage is that some judgments about language are sanctioned by tradition – a tradition not of using language but of judging it. Ending a sentence with a preposition, for example, has been a long-enduring and widely known proscription, and the tradition itself has been instrumental in maintaining the proscription (see also Reference CurzanCurzan 2014, 87–89). Traditional usage could be a subset of good usage, since all the traditional prescriptions are (at least initially) claimed to constitute good usage, and for that reason dictionaries seldom appeal to traditional usage explicitly, but they may acknowledge traditional usage whenever they refer to the tradition of rules, as when Reference PickettAHD (2011) writes about hopefully that its use as a sentence adverb, as in Hopefully, the measures will be adopted, “seems to have taken on a life of its own as a sign that the writer is unaware of the canons of usage.” Traditional usage deserves to be discussed separately from good usage partly because of a separate sense of usage that turns out to be key and partly because of its importance in prescriptive practice.

The separate sense in traditional usage is “customary practice linked to tradition.” Reference WorcesterWorcester (1860) links tradition to usage with the key notion “received”: “Practice long continued; received practice; custom; use; habit” (s.v. usage). In this sense, usage refers to some tradition that recognizes and approves a specific practice. That such a practice can be in domains other than language comes out in Webster’s New International Dictionary, Second Edition (1934): “the prevailing mode of procedure (as of a craft, business, liturgical tradition) […] the chapel services follow the usage of the Episcopal church” (s.v. usage). A tradition thus establishes a correct or standard way of doing things. The Concise Oxford Dictionary explicitly includes this standard-creating sense: “habitual or customary practice, especially as creating a right, obligation, or standard” (Reference PearsallPearsall 1999). When it comes to words, then, usage can refer to the creation of a standard by the operation of some tradition. The Funk & Wagnalls Standard Dictionary of the English Language states this sense explicitly: “A use of words or forms considered as sanctioned or unsanctioned by reputable authorities” (Reference MarchMarch 1893–1895). So in this sense, usage does not refer to how words are actually used (i.e. actual usage) but instead to words that conform to the judgment of “reputable authorities” (i.e. traditional usage).

In addition to these definitions, prescriptive practice has also assumed a connection between usage and tradition, sometimes explicitly, as when the term “traditional rule” is used, but more often implicitly. The notion of tradition is crucial to the economy of prescriptive advice, which promises that speakers and writers will be rewarded for learning prescribed forms and punished for using proscribed forms. A usage guide from 1942 states this assumption bluntly:

For readers to profit from such prescriptive advice, the constructions that distinguish the literate, educated, and refined from the illiterate, uneducated, and vulgar have to be identifiable and manageable so that a person can learn them and avoid the finger of scorn.

The rules also have to be recognized by enough people for the claimed rewards to inhere. Essentially, some manageable number of rules is needed for judges to recognize and for aspiring students to learn. Such rules act as shibboleths, revealing who has taken the time to learn these specific rules and subjected themselves to the discipline of using the prescribed variants. As Reference LandauLandau (2001, 256) says, “When we talk about disputed usages, or what I should rather call ‘class markers,’ we are not marking the class of those who use them so much as those to whom the usage is objectionable.” Items like adverbial quick, verbal impact, or backformed enthuse are not necessarily “nonstandard,” as they are used in standard English, both spoken and written, in ways that double negatives, double modals, and ain’t are not.

The important work that such proscribed forms do is to signal those who know to object to them. Shibboleths like these identify those who know the rules and care about following them (Reference Chapman, van Ostade and PercyChapman 2017). As Dwight Reference BolingerBolinger (1980, 168) pointed out, learning the rules is costly, and that is what makes them useful as a “social password.” This shibboleth function of rules – dividing the knowledgeable from the ignorant – is what Laurie Reference BauerBauer (1997, 9) describes when he says, “If everyone learnt to use imply and infer consistently, some other prescription would arise to distinguish between the in-group and the out-group, because making that distinction is an important function of language.” The use of certain prescriptive rules to mark in-groups and out-groups necessitates a tradition that can perpetuate the social meaning of those shibboleths.

Steven Pinker uses the notion of “common knowledge” to capture the practice of shibboleths. Common knowledge is not just knowledge that is common to many people, but knowledge that people assume is common to all. Usage rules are treated as common knowledge, in that we assume that there are some language rules out there that everyone knows, and if we don’t know them, we will be left out. That is why we need the advice. But these rules must be known and knowable by enough people for us to trust that they are common knowledge (Reference Pinker2011, xvii). That is why a shibboleth cannot be just any instance of linguistic variation. It has to be variation that people can readily know or find out about. The operation of tradition in traditional usage follows from the promises of learning good usage.

Traditional usage in dictionaries is most apparent in the deployment of usage notes. Dictionaries do not provide usage notes for all words because not all words signal the in-group speakers who have learned The Rules. Instead, dictionaries focus on those rules believed to be common knowledge. In principle, those four or five hundred entries that do contain usage notes in a general dictionary would constitute a group small enough to be learned and recognized. Indeed, certain words (e.g. irregardless, hopefully, and flaunt vs. floutFootnote ⁴) come with usage labels or usage notes in multiple dictionaries,

Such commonly repeated prescriptive rules, important as they are in the prescriptive tradition, have not had a single term to identify them. I have been using shibboleth, which partially fits but may have too strong connotations. Another term increasingly used is “old chestnuts” (Reference Tieken-Boon van OstadeTieken-Boon van Ostade 2020, 105–108). Some scholars have written about a loosely defined canon of well-known rules within the prescriptive tradition (Reference NunbergNunberg 2000, xix). Unfortunately, there is no systematic way to identify “old chestnuts,” much less a canon of usage rules that are widely known and sanctioned by the tradition. To be sure, certain rules are frequently repeated in dictionaries and usage guides, but there are none that are always repeated. There is no strong convergence among usage commentators.

This brief survey of usage has shown that good usage – the simple dividing of variants into good and bad or correct and incorrect judgments – is not the only kind of usage included in dictionaries, and even good usage is related to actual usage, contextual usage, and traditional usage in that it can be supported by these separate but overlapping authorities. Good usage can be ascertained by empirically noting how often a construction is used or the contexts that it is used in or by noting the tradition of authorities promoting or deprecating it.

General dictionaries are in a reasonably good position to provide all three types of evidence for good usage. Their empirical methods will give editors a relatively clear idea of actual usage and contextual usage, even though the dictionary format would normally preclude presenting quantified evidence. Thus, notwithstanding its frequent proscription, a dictionary can (and most do) include the ‘conducive to health’ sense of healthy because the empirical record readily documents that sense. Some dictionaries add a usage note to address the traditional usage separate from the actual usage, but few exclude the proscribed, though well-attested, sense.Footnote ⁵ Additional methods also allow dictionaries to pronounce on traditional usage, since most editors check other general and specialized dictionaries. Such consultation would give them evidence for what usage tradition has sanctioned. And general dictionaries are in a good position to give authoritative pronouncements for items that are expected to be proscribed authoritatively. In short, they constitute an outstanding platform for pronouncing on good usage, and they use reasonably good methods for gathering evidence for those pronouncements. If the question is whether a given sense like ‘reluctance’ for reticence or ‘devastate’ for decimate is good usage, general dictionaries are in a reasonably good position to answer that question and in an even better position to be trusted.

But general dictionaries are in a less advantageous position to answer many other questions about usage. If, for example, the question is of which debatable senses of words one should be aware, a general dictionary is not as good a resource, since the debatable senses are scattered throughout the dictionary. Similarly, if the question is which grammatical variants are proscribed, such dictionaries are less than optimal because of their focus on lexis instead of grammar. In short, general dictionaries are reference books, so when the question is a reference question, like what is the usage judgment of a specific word, dictionaries are a good resource. But when the information requested is how to apply usage judgments, they are not so good (see Reference LandauLandau 2001, 263). For the how-to information, a specialized dictionary – the dictionary of usage – has arisen.

16.3.1 Titles and Generic Names

Books providing usage advice go by a variety of generic names besides “dictionaries of usage,” including “usage books,” “usage handbooks,” and “usage manuals” (Reference CreswellCreswell 1975). Recently Ingrid Tieken-Boon van Ostade and scholars involved with her “Bridging the Unbridgeable” project have consistently used the term “usage guides” (see especially Reference StraaijerStraaijer 2018 and Reference Tieken-Boon van OstadeTieken-Boon van Ostade 2020). The terms that usage guides use in their own titles are just as varied. Books aimed at giving practical advice often deploy catchy titles emphasizing self-help, like Get It Right! A Cyclopedia of Correct English Usage (Reference OpdyckeOpdycke 1941). Shorter, mass-market books may also adopt humorous titles like Woe Is I (Reference O’ConnerO’Conner 1996) or Lapsing into a Comma (Reference WalshWalsh 2000). Guides that are framed as authoritative reference books usually have more staid names. Many call themselves a dictionary, as with Fowler’s Dictionary of Modern English Usage (Reference Fowler1926), Merriam-Webster Dictionary of English Usage (Reference Ward Gilman1989), Dictionary of Contemporary American Usage (Reference Evans and EvansEvans and Evans 1957); others call themselves a guide, as with The Cambridge Guide to English Usage (Reference PetersPeters 2004), The Columbia Guide to Standard American English (Reference WilsonWilson 1993), and Modern American Usage: A Guide (Reference FollettFollett 1966). Reference GarnerGarner (2022) confidently omits dictionary and guide in his Garner’s Modern English Usage. Most of these large, reference books also use the term usage, though a few use some other term for approved English, like standard. They often include the name of their reputable publishing company in the title, as with Columbia (Reference WilsonWilson 1993) or Cambridge (Reference PetersPeters 2004). If there is a difference between a dictionary of usage and a usage guide – and I am not aware of anyone claiming such a difference – it might be that a dictionary of usage is a large, reference authority, while a usage guide would be the overarching term for all books treating usage. This distinction is not widely recognized and does not have to be signaled in a book’s title. I prefer usage guide as the generic term, though I have used it nearly interchangeably with dictionary of usage and usage dictionary in this chapter.

16.3.2 Selection of Usage Items

The items treated in dictionaries of usage vary widely from dictionary to dictionary. Robin Straaijer and Ingrid Tieken-Boon van Ostade have examined this divergence using the HUGE database they created, which surveys the treatment of 123 usage items across seventy-seven usage books (Reference StraaijerStraaijer 2014). Reference Tieken-Boon van OstadeTieken-Boon van Ostade (2020, 106) notes that not one of the seventy-seven books contains all 123 items in the database and that the most convergence was seen with shall vs. will, which was included in sixty-five of the books. Reference StraaijerStraaijer (2018, 27) points out that Reference BurchfieldBurchfield (1996) includes the largest number of these usage items, at 111, but that the range across the books is wide, as Reference HorwillHorwill (1935) includes only eight of the 123 items. I have found similar trends in my own unpublished survey of all rules in thirty usage guides.

This divergence is surprising given the rewards promised for learning prescriptive rules. A high priority for usage guides would presumably be to identify those shibboleths that most often and most reliably distinguish speakers and writers who know and follow The Rules. Yet such identifying is hard to find in usage dictionaries, where the rules are given without hierarchy. Within individual entries editors may often note a rule’s relative shibboleth status, but I am not aware of any usage dictionary structured to signal such status within distinct sections or with an index. When usage dictionaries include thousands of entries, identifying the most important rules would be impossible without such signaling. Cambridge treats over 4,000 items, Garner over 5,000, and Fowler over 7,000. Much as general dictionaries have a needle-in-a-haystack problem with their 400 to 600 usage notes spread out among thousands and tens of thousands of entries, large usage dictionaries have their own needle-in-a-needlestack problem in separating the most important rules from all the others. Even the smaller usage dictionaries that treat a more manageable number of usage issues do not help, since they do not converge on a canon of shibboleths either. In my own sample of seventeen usage guides with fewer than 1,000 items, only twenty-six items occurred in more than half of the guides, and only 634 occurred in more than one. Taken as a group, they treated 2,237 different items, still too large a number for a canon. Roy H. Reference Copperud and CopperudCopperud (1980, vi), who surveyed nine usage dictionaries in order to find consensus, succinctly states the problem: “The implication is that the critic is reflecting the preponderance of educated practice. But this is not necessarily so, or there would be more agreement among the authorities.”

Yet it appears that editors exercise wide latitude in their selection without necessarily realizing it. Reference Tieken-Boon van OstadeTieken-Boon van Ostade (2020, 108) reports one editor telling her that he “aimed to present popular usage problems,” and Anne Reference CurzanCurzan (2014, 76) reports that the group implementing the Microsoft Word grammar checker “used a combination of popular style guides.” Editors apparently assume that their collection is based on some sort of collection of traditional rules, yet they have not tightly defined such a collection. At the same time, editors have also revealed that their own purposes and predilections informed their selection. One editor reported that his purpose was “to write about more basic things – particularly how those officials could make their business documents […] clearer for the mass of people,” and at the same time he “wanted to lay down a personal marker to help make [his] name in the field” (Reference Tieken-Boon van OstadeTieken-Boon van Ostade 2020, 226). Another editor reported that “I looked at several more serious and comprehensive guides to English and chose examples […] that appealed to me. […] But it was very personal – I basically wrote about the things that amused or annoyed me” (Reference Tieken-Boon van OstadeTieken-Boon van Ostade 2020, 230). Guides have their own personalities and predilections. Reference BernsteinBernstein (1965), for example, has more word complementation issues (like take heart from/at), Reference Evans and EvansEvans and Evans (1957) more synonym discrimination entries (like delightful; delicious; delectable), Reference GarnerGarner (2022) more technical legal terms (like acquittal; acquittance; *acquitment), and Reference FowlerFowler (1926) more French words (like bienséance) than others. So one reason that usage guides do not converge on the usage issues they treat is that they allow for a fair amount of the editors’ personal preferences.

Another reason for the divergence is that usage dictionaries use criteria that extend beyond identifying the most important rules for speakers and writers to know. Bryan Garner lists ten principles that guided his selection, and only a couple of them explicitly mention avoiding negative judgments. Most of his principles highlight ways to make writing clearer or more effective (Reference GarnerGarner 2022, x). These rewards are slightly different – a writer may be considered a better writer for following the advice, but not necessarily uneducated, illiterate, or vulgar for not following it – and these lower stakes make it possible to include more types of variation than those that purportedly identify correct and incorrect use. These additional types may include choices that distinguish between more and less precise use of words, briefer and wordier phrasing, or more and less fitting expression. These distinctions correspond roughly with Curzan’s “stylistic prescriptivism” as opposed to her “standardizing prescriptivism” (Reference CurzanCurzan 2014). Practically any distinction that an editor can recognize could be included in a usage guide, since such distinctions are claimed to make the language more capable of precise expression (Reference FollettFollett 1966, 10). The number of potential usage rules is already high since they are essentially judgments about variation and variation is ubiquitous in languages. Extending the judgment of variation to rhetorical considerations makes the potential number of usage rules very high.

16.3.3 Dictionary Structure

The structure of usage dictionaries also varies widely (Reference StraaijerStraaijer 2018, 23). A few like Reference 680HefferHeffer (2010), Reference LambLamb (2010), and Reference TaggartTaggart (2010) display a discursive presentation, but most have packaged their advice into discrete entries for each usage issue. The earliest usage guides like Reference BakerBaker (1770) and Five Hundred Mistakes (Reference Burgess1856) provided no discernible order to their entries, but most guides since have ordered their entries in some way. Some have placed their entries into topics, like “Puzzling Plurals” or “Problem Prepositions” (Reference BatkoBatko 2004) without any further order, while others, like the American Heritage Book of English Usage (1996), have ordered their entries alphabetically within topics. Books using this macrostructure often include a general index to their usage issues. The most popular structure in twentieth- and twenty-first-century usage dictionaries, especially in the large dictionaries, has been alphabetical order of discrete entries without division into topics.

16.3.4 Entry Names

A large challenge for usage dictionaries is choosing names that will be meaningful for the variety of people searching for usage advice. Most usage issues involve at least two forms – a prescribed and a proscribed form – and using both forms in the headword has become typical, as with affect / effect, imply / infer or different from / different than. Morphology can complicate the choice a little – should the name be emigrate / immigrate or emigrant / immigrant? should the name be aggravate or aggravation? – but a much larger complication comes when a grammatical or linguistic category is more important than single words. In grammar, for example, several usage issues involve subject–verb agreement, such as whether the verb should be is or are after either of them. Should this issue be treated in an entry named after two forms, as possibly either of them are / either of them is? Or should it be treated in an entry named after the function, say subject–verb agreement with perhaps a subtype for either? Similarly, a spelling issue is whether to use two letters or one before verb endings, as in travelled or traveled. Should this issue be addressed in all applicable verbs, like fellowshipped vs. fellowshiped? Or should it be treated in a single entry named after the more general issue, perhaps labeled double letters before verb endings?

Dictionaries seem to have gravitated toward offering lexical entry names wherever they can, sometimes with cross-references to longer articles explaining the more abstract categories. Thus Reference GarnerGarner (2022) includes entries named it is I; it is me for case following a copular verb, between you and me; *between you and I for conjoined object pronouns, and than whom for cases following than, as well as a general article named Pronouns. It is obviously difficult to predict what name a reader will think to use in a search, and efforts to find the best lexical and category names have led to different names for the same usage issue.

The use of lexical names, even for category problems, nonetheless makes sense, as it is the simplest way for naive readers to find usage information; they do not have to know the analysis or metalanguage of the construction before they can find out whether, say, everyone … they is “correct.” This naming practice is part of a larger trend toward greater lexical treatment of usage rules. A plurality of usage items in most usage guides are explicitly based on word meaning. Some items, for example, proscribe senses for individual words, like the intensifier sense of literally, the permission sense of can, the problem sense of dilemma, and the devastate sense of decimate. Others try to differentiate the senses of two similarly shaped words, like effectively vs. effectually, disinterested vs. uninterested, transcript vs. transcription. Still others aim to keep near synonyms distinct from each other, like lady vs. woman vs. female, deem vs. think, lawyer vs. attorney vs. barrister vs. counsel. Still others keep word meanings matched with their right spellings in homophone and near-homophone pairs, like they’re vs. their vs. there, it’s vs. its, or effect vs. affect. Beyond these denotative meanings, usage items also treat the contextual usage considerations of individual words, like their currency and formality. Still more usage rules treat issues seen as a property of a word, like the spelling, abbreviation, pronunciation, plural form, and past tense forms. Similarly, the complements that a word takes (different from vs. different to) can also be seen as a property of a word, and therefore lexical. Furthermore, the idioms and fixed phrases that some dictionaries treat are more lexical than grammatical. When all such issues are added up, they overwhelm those that would not be lexical, like grammar (not including lexico-grammar) and punctuation. In my own sample of 5,755 items, 5,265, or 91 percent, are lexical in one of the ways just listed. The remaining 9 percent are not lexical, but they are usually given lexical headwords wherever possible.

16.3.5 Entry Structure

The structure of entries varies widely within and across usage dictionaries, as editors decide differently on how much information to include. A few of the most important and most variable features of entry structure are noted here.

18.2.1 Sources and Models: The Pre-Modern Era

Throughout the first 1,500 years of biblical study, countless handwritten resources appeared that passed down the centuries, providing both sources and models for later dictionaries produced in the print era. Aside from copies of the biblical texts themselves, the most important of these were Greek and Hebrew wordlists of various kinds, which contained valuable data. But other types of ancient and medieval reference works also proved as influential in form as they were in content.

The main precondition for the emergence of these early resources was the Hellenistic philological tradition, mostly centered in Alexandria in the third through first centuries BCE, which laid long-lasting foundations for the later European humanities. The Greek term φιλολογία (philologia) itself had both linguistic speculation and the study of rhetoric within its purview, but it also referred to detailed textual and grammatical analysis focused on Classical texts. These latter tasks in particular involved the production of scholarly works such as grammatical treatises and several types of monolingual glossaries. The more lexically oriented works tended to focus on clarifying difficult terminology used in the works of a single author like Homer or Hippocrates, or declaring which Attic vocabulary was acceptable among the high-minded literati of the Second Sophistic and how best to use it in the first and second centuries CE (Reference Dickey, Montanari, Matthaios and RengakosDickey 2015; Reference 753Tosi, Montanari, Matthaios and RengakosTosi 2015).

It was this Greek philological tradition – enmeshed in the later Roman culture – from which early Christian scholarship took its cue, but with important advancements. Most relevant here is the activity of Eusebius of Caesarea (c. 260–339). As bishop in a newly Christianized empire, Eusebius enjoyed decades of lavish sponsorship and access to a massive library, which he leveraged to compile numerous anthological tomes. For example, his Chronicon recounted world history from Abraham to Constantine, carrying on earlier Greek forms of antiquarian scholarship but with unprecedented comprehensiveness and a revolutionary organizational strategy for presenting information visually on the page.Footnote ⁷ Another of his key works, the Onomasticon, provided a directory of biblical toponymy, topography, and ethnography, organized by headword – alphabetically by first letter only – in the tradition of earlier wordlists (Reference FlandersFlanders 2020, 25–26). In the course of his work, Eusebius collated a great deal of information germane to untold generations of biblical interpreters, earning him high praise even from such towering interpreters as Augustine (PL 34:62). He also hammered out new methods for organizing and presenting it that foreshadowed Bible dictionaries produced over a millennium later in content, structure, and format (Reference Grafton and WilliamsGrafton and Williams 2006, 133–177).

While Eusebius’ work remained influential for centuries, the felicitous conditions that made his scholarship possible were far less enduring. Beginning in the fifth century, major geopolitical events put new and long-lasting boundaries into place in religion, culture, and language, dramatically limiting the kind and amount of information available to corral into reference works for biblical interpretation. First, the fall of the western Roman Empire cleaved Greek-speaking scholarship in the Byzantine (later Orthodox) East from the Latin-speaking European (later Catholic) West. Second, the rise of the Ummayad Caliphate in the seventh century in the Iberian Peninsula and beyond established an environment for Jewish scholarship that was mostly hospitable – especially in the tenth and eleventh centuries – and therefore mostly separate from the interpretive milieu of the Church. In each of these regions, new scholarly reference works appeared and old ones were passed down, but they largely remained within religiously and linguistically reinforced interpretive silos.

On the one hand, the reference works produced in Byzantium that would prove most important for later biblical interpretation were not actually specific to the Bible itself. Because Eastern scholarship and education persisted in the Hellenistic pattern, a robust Greek lexicographical tradition continued there. The crown jewel appeared in the Suda, a tenth-century Greek dictionary with around 30,000 headwords covering a wide array of ancient literature, organized alphabetically and briefly defined or discussed, mostly on the basis of lost Greco–Roman lexicons (Reference 713DickeyDickey 2007, 87–106; Reference ValenteValente 2019). In this same period, Jewish scholars were much more directly focused on their sacred texts, including the Talmud and Midrash alongside the Bible. The apex of medieval Jewish philology occurred from the tenth through twelfth centuries. This tradition developed in a multilingual context and with the exposure to Arabic grammarians necessary to foster it. A vibrant culture of Hebrew lexicographical scholarship sprang up, producing well-edited wordbooks for biblical Hebrew. These included the Egron by Saʿadia Gaon (c. 882–942) in the East and the Maḥberet by Menaḥem ben Saruq (c. 920–70) and the Kitāb al-ʾUṣūl by Yona ibn Janāḥ (c. 990–1055) in Al-Andalus, not to mention the bilingual glossaries that became available in several European vernaculars (Reference DelgadoDelgado 2013; Reference de Langede Lange 2013; Reference MamanMaman 2013 and Reference Maman2019; Reference Sáenz-Badillos and de LangeSáenz-Badillos 2001).

On the other hand, Western biblical scholarship was more a matter of survival in the medieval period. Knowledge of Greek and Hebrew was exceptionally sparse among Christians, reaching a nadir by the Fourth Crusade and establishment of the Latin Empire (1204). Traces of rudimentary Hebrew do appear in the writings of the Englishmen Bede and Alcuin of York, likely owing to their appropriation of the work of Isidore of Seville (c. 570–636), whose encyclopedic and widely influential Etymologiae continued in the Eusebian antiquarian tradition over twenty-one volumes (Reference JonesJones 1983, 8–9; Reference Merrills, König and WoolfMerrills 2013). Hebrew gained more attention from the twelfth century, aided by private Jewish instruction. Figures such as Hugh of St. Victor (c. 1096–1141) and Roger Bacon (c. 1214–1275) saw value in the language for correcting the text of the Vulgate, spiritual exegesis, and non-coercive Jewish proselytizing (Reference JonesJones 1983, 9–13; Reference Olszowy-SchlangerOlszowy-Schlanger 2013). As for Greek, the proud few who had some competence in the language were concentrated mostly in the eighth and ninth centuries during the Carolingian Renaissance, subsisting with only whatever bilingual glossaries were to hand, usually those surviving from antiquity or drifting in from the East, and not specific to the biblical corpus. These rare tools nevertheless allowed some western scholars to get sufficient grasp upon the language to compare Greek and Latin in parallel columns or interlinear texts (Reference Herren, Archibald, Brockliss and GnozaHerren 2015; Reference Rotolo and ChristidisRotolo 2015).

Of course, Latin was the medium of Western scholarship, and a range of reference work genres appeared in Latin that could be used to interpret both the language and content of the Bible. As for language, the monolingual glossary was again very prominent, often the result of collecting various marginalia elucidating the meaning of rare words in literary, patristic, and biblical texts. The seventh century Liber Glossarum, for example, presented almost 30,000 Latin headwords (drawing selectively upon Isidore’s Etymologiae). That these were listed in near absolute alphabetical order – an arduous, expensive, and still innovative approach – indicates the munificent cultural support of Latin scholarship compared with the other biblical languages (Reference ConsidineConsidine 2019b, 268–272). Aside from more language-oriented reference works, medieval scholars also set the mold for numerous types of resources for epitomizing and collating the content of Scripture and related writings, both ancient and contemporary. This instinct likely goes back to Aristotle’s suggestion to organize information under subject-based headings (Top. 1.14). But it carried on in the medieval West in a variety of genres, including chronologies, collationes, and florilegia, among others, the desirability of which grew in connection with Scholasticism and the renewed emphasis on historically grounded exegesis as championed by Hugh of St. Victor (Reference FlandersFlanders 2020, 43–70; Reference TurnerTurner 2014, 25–32; and Reference van Lierevan Liere 2014, 141–176).

18.2.2 Collection and Elaboration: The Early Modern Era

At the risk of crass reductionism, the most important outcomes of the first 1,500 years of biblical scholarship were vast heaps of basic linguistic and textual data, as well as strategies for organizing that information on the page in logical and quickly accessible ways (Reference McKitterick, Kwakkel and McKitterickMcKitterick 2012). To say that the benefit of these outcomes was pragmatic for later scholars is not to diminish their importance in the history of the dictionary in general and its influence in broader society in the early modern period and beyond.Footnote ⁸ To the contrary, it would be all too easy to overlook these and other logistical necessaries of dictionary-making that arose in the medieval era, often in connection with biblical interpretation. The arrival of paper and Arabic numerals via Islamic Spain, for example, made textual production cheaper and facilitated inserting chapter and verse numbers into Scripture (and other ancient literature), which in turn led to the invention of the Bible concordance (Reference BloomBloom 2001, 125–213; Reference FlandersFlanders 2020, 71–90).Footnote ⁹ It was these sources, models, and technologies that biblical interpreters in the Early Modern era eagerly inherited, refined, and enhanced, but now with the terrific adaptability and speed of movable type and the printing press.

As in late antiquity, major geopolitical changes transpired in the late fifteenth century that shifted the landscape of possibilities for reference works. Following the fall of Constantinople to the Ottomans in 1453, the Alhambra Decree issued in 1492 by the Spanish monarchy, and the onset of the Habsburg-Valois wars in 1494 by Charles VIII of France, European scholarly culture was infused with the very intellectual communities from which it had been mostly separate for a millennium. Waves of Greek, Jewish, and Italian philologist-refugees fled their homelands, bringing their texts and language expertise with them and thus centralizing the resources of Renaissance learning in the territories northwest of the Alps. It was within this early modern intellectual milieu that the bilingual lexicon and the Bible dictionary – two of the three categories of dictionaries discussed above – crystalized into their more readily recognizable profiles.

Early modern Hebrew and Greek lexicography built directly upon the foundations laid in Byzantium and Al-Andalus and lately imported, but now using the prototypical dictionary framework that was modeled in the 1502 Latin Dictionarium by Ambrogio Calepino. This dictionary was highly influential, in part because it was imperfect. After considering a revision, the French humanist, biblical scholar, and printer Robert Estienne set out instead to improve upon Calepino’s Dictionarium, which culminated in the monumental Linguae Latinae thesaurus in 1531 (Reference ConsidineConsidine 2019a, 292–293). That same year saw the birth of another lexicographical tour de force: Robert’s son and philological heir, Henri. After Robert’s death thirty years later, Henri assumed the task of completing his father’s comprehensive dictionary of ancient Greek, study of which had revived in the West in the preceding century (Reference Kraye and CameronKraye 2016). Henri’s lexicographical magnum opus was published in 1572 as Thesaurus Grecae linguae, covering over 60,000 words and modeled in both practical and theoretical terms upon the two preceding Latin dictionaries (Reference ConsidineConsidine 2008, 67–100). Although it covered ancient Greek in general, rather than New Testament Greek in particular, the significance of Henri’s Thesaurus for biblical interpretation is difficult to overstate. In 1579 Henri’s assistant, Johannes Scapula, produced a one-volume abridgement of the Thesaurus, which was followed in 1594 by Henri’s Concordantiae testament novi graecolatinae. Both of those works in turn fed directly into the first substantial New Testament lexicon, produced by Georg Pasor in 1619. It was Pasor’s lexicon – often reprinted, revised, and translated – that in turn provided a foundational model and dataset for New Testament lexicons over the next three centuries or more (Reference Lee and CarsonLee 2003, esp. 61–81).

Lexicons of biblical Hebrew continued to appear and improve within Jewish scholarship into the early modern period too, particularly building on the Sefer ha-Shorashim by David Kimḥi (c. 1160–1235). Kimḥi’s own work relied in turn upon Yona ibn Janāḥ and spread widely throughout Europe (Reference MamanMaman 2019, 193). But an important cultural shift occurred in the sixteenth century. A movement known as Christian Hebraism grew out of the soil of Renaissance humanism, cultivated by theological developments during the Reformation. The motivations for Christian Hebraism were multilayered, but of principle importance was a desire to evangelize Jews – sometimes itself underwritten by cataclysmic eschatological expectation – or to engage in Protestant–Catholic theological polemics concerning the locus of authority: biblical text or papal magisterium (Reference CampaniniCampanini 2013, 442; Reference TurnerTurner 2014, 39–44). With such ends in mind, Christian Hebraists collected and repurposed resources produced by Jewish scholars themselves, ultimately prompting the creation of royally funded chairs of Hebrew at several European universities (Reference BurnettBurnett 2012). A key figure in this regard was Johannes Reuchlin, who in 1507 revised Kimḥi’s work in Latin under the title De rudimentis Hebraicis. It included the first of more than seventy Hebrew lexicons for non-Jewish readers to be printed before 1800, the most significant of which was no doubt Johannes Buxtorf’s 1607 Lexicon Hebraicum et Chaldaicum, reprinted over twenty times right into the modern era (Reference Jones and GreensladeJones 1963, 523–524).

As Christians gained hold of Greek and Hebrew, currents in biblical interpretation began to cross in new ways with other areas of early modern society and culture, giving rise to the Bible dictionary as a scholarly desideratum. As Scripture became more widely and cheaply accessible to laity in vernacular languages, it profoundly destabilized the authority so long centralized within the matrix of Western political and religious institutions. Few in the sixteenth and seventeenth centuries rejected the authority of Scripture outright. Who could gain access to that authority (and how) was quite another matter, however, particularly as interpretive traditions split and multiplied. Interpretation came to center stage.Footnote ¹⁰ For Protestants, the Bible alone held authority, not only for Church doctrine but also for matters of natural philosophy. So best practice for navigating between the Scylla of interpretive anarchy on the one hand and the Charybdis of Catholic dogma on the other commended jettisoning the ancient and often freewheeling allegorical interpretive method in favor of a historically grounded “literal” sense of the sacred text.Footnote ¹¹ Any effort to capture that hermeneutical phantom thus required martialing as much information about the biblical text and its authors as possible – the value of language, geography, law, politics, realia, and more were now appreciated as never before. The discovery of new worlds, recovery of ancient texts, and proliferation of printed books only increased scholarly fervor in this regard. This impulse among early modern biblical interpreters was both religious and humanistic in character. As such, it entailed the same array of antiquarian and Orientalist efforts to reconstruct the distant past as those made by European scholars studying other ancient texts – all of whom were buoyed by the notion of historia sacra – and it resulted in the compilation of the voluminous results into printed formFootnote ¹² (Reference Blair, König and WoolfBlair 2013, esp. 381; Reference Harrison and CameronHarrison 2016; Reference MillerMiller 2001; Reference Rein and CameronRein 2016; Reference TurnerTurner 2014, 44–56).

Among the new reference work genres to arise from the tireless encyclopedism of the early modern period was the Bible dictionary. A good sense for the rationale behind such a resource appears (conspicuously) in a review of a mid-eighteenth-century New Testament Greek lexicon. The anonymous reviewer reminds the reader that

Perhaps ironically, the urge to meet such needs was not lost upon the author of the lexicon himself, the English clergyman and scholar John Parkhurst, who in his lexicon entries tended to indulge in prolix digressions. For example, in the entry δοῦλος (‘servant, slave’), Parkhurst cites secondary literature to elaborate upon the “deplorable” conditions of “these unhappy persons.” He then clarifies that he “transcribed the above affecting account of slavery according to the Roman law, because by it we shall be the better enabled to enter into the full meaning and spirit of several passages of the N. T.” (see also Reference Lee and CarsonLee 2003, 92–96). This lexicographical apology and the comments by Parkhurst’s reviewer help to triangulate the position in which early modern scholars found themselves, one that ultimately gave rise to the Bible dictionary as a comprehensive yet navigable resource for their interpretive endeavors.

18.2.3 Theory and Synthesis: The Modern Era

Though it took off especially in the nineteenth century, in reality the Bible dictionary genre first appeared at the beginning of the early modern era. The foundational work had been that of the French Benedictine monk Antoine Augustin Calmet, whose four-volume Dictionnaire historique et critique, chronologique, géographique et littéral de la Bible was complete by 1730.Footnote ¹³ Within two years it had been translated into English by Reference d’Oyly and ColsonSamuel d’Oyly and John Colson (1732), with a sprawling title describing its scope, content, and even structure. It reads (in part!) much like the list by Parkhurst’s reviewer:

This Bible dictionary was then emended, revised, and reprinted by Charles Taylor in 1795, and went through five editions by 1829, by which time it was “augmented” even further “with an extensive series of plates, explanatory, illustrative, and ornamental.” Taylor’s edition was then brought to American shores, where it was revised again in 1832 by Edward Robinson (who appears again below), with an abbreviated version printed in 1833 as A Dictionary of the Holy Bible. A great many others followed in various languages, some of which appear in the discussion below, others branching off as (en)cyclopedias, and many with some level of indebtedness to Calmet (see Reference DankerDanker 1993, 148–153; Reference Gottheil and SingerGottheil 1902; Reference HorneHorne 1839, 369–372).

While biblical lexicons remained indispensable, with the rise of the Bible dictionary in the nineteenth century the turn to historical knowledge as the sine qua non of biblical interpretation was complete. Of course, when it comes to producing a Bible dictionary, someone has to select what goes into such a supposedly exhaustive and authoritative resource and what does not, a task that comes with unusual power given the cultural status of Scripture itself. To be sure, handling some topics is fairly straightforward – the location of rivers, names of gemstones, and types of coins, for example. But to navigate other topics – say, identifying and explaining the “Habits” and “Curiosities of the Jews” as in the title of d’Oyly and Colson’s Dictionary (Reference d’Oyly and Colson1732) – is to tack and jibe through the headwinds of contemporary thought, whatever they might be. It is no surprise, then, that the Bible dictionary has both dissimulated and promulgated the ideologies of contributors or editors, whether to challenge or reinforce otherwise socially acceptable views. The first case study below begins to bear this out. By the same token, the intellectual conditions that gave rise to the theological dictionary are complex, tied elaborately to the robust mast of nineteenth-century comparative historiography in matters of language, religion, cultural development, and beyond. But that discussion must await its turn in the second case study.

These same complex conditions enveloped biblical lexicography in the modern era as well. In the mid-eighteenth century, James Harris revived the Aristotelian view that language was a reflection of thought, such that the features of a language were taken to reflect the intellectual and cultural characteristics of those that speak it. In his Hermes, Harris purported to show how “Nations, like ſingle Men, have their peculiar ideas; how theſe peculiar Ideas become the Genius of their Language, ſince the Symbol [language] muſt of courſe correſpond to its Archetype [thought]” (1751, 407). In the nineteenth century, scholars increasingly applied this view to the language of the New Testament, which for centuries had been debated as somehow influenced by Hebrew in syntax or semantics. New Testament Greek, many in the eighteenth and nineteenth centuries argued, was totally distinct from the broader language system of its time; it was categorized in turns as a unique “Holy Ghost Greek” or simply as a degraded and inferior offshoot of “proper” Greek. For some, the best explanation of its peculiarities was, of course, that it was Jewish, a product of “men whose thoughts were cast in a Semitic and not in a Hellenic mould” (Reference HatchHatch 1889, 10).

In reality, however, the Greek of the New Testament (and of the Septuagint) looked peculiar largely due to the fact that it was being compared with the wrong thing: Classical Attic. Greek scholarship was ushered into a radically new era at the turn of the twentieth century, as massive troves of nonliterary, post-Classical Greek papyri were excavated in Egypt, revealing a much more nuanced linguistic picture. Now apples could be compared with apples in diachronic and sociolinguistic terms. And that is precisely what began to happen, particularly in the work of Adolf Reference DeissmannDeissmann (1895), who demolished the “Jewish Greek” hypothesis, even if it took scholars some time to notice (Reference Horsley and HorsleyHorsley 1989). The shift in scholarly outlook was by no means immediate or comprehensive. But discovery of the papyri set New Testament lexicography on a new heading, leading to the groundbreaking lexicon by Reference Moulton and MilliganMoulton and Milligan (1929), the first to make any systematic attempt to account for papyrological evidence in its semantic analysis of New Testament Greek (Reference Lee, Taylor, Lee, Burton and WhitakerLee 2004; Reference TurnerTurner 2014, 85–88 and 140–146; Reference Voelz, Temporini and HaaseVoelz 1984, 900–919).

Hebrew lexicography was not left unaffected by these developments, particularly as comparative philology swept over European scholarship at the turn of the nineteenth century. This new method had deep roots in Germany, especially the Hebrew scholarship of Albert Schlutens (1686–1750) and Johann David Michaelis (1717–1791). The latter was professor at Göttingen for over forty years, where he trained the father of modern Hebrew lexicography, Wilhelm Gesenius. It was Gesenius’ epoch-making work, the Hebräisch–deutsches Handwörterbuch (1810–1812), that finally dethroned Buxtorf’s Lexicon and subsequently went through countless reprints, revisions, translations, and adaptations right up to the present (Reference HoltzHoltz 2013). As shown in the first case study, Gesenius became – as it remained until the late twentieth century – the default, authoritative Hebrew lexicon for interpretive arbitration, even when it was viewed with a measure of suspicion for culturally embedded reasons, as appears below.Footnote ¹⁴

18.3.1 Dictionaries Upstream: Biblical Interpretation and Slavery in America

In a fundamental sense, dictionaries are tools. But, like all tools, they are only as effective as the quality of their craftsmanship and the competence of their user permit. Not only that, but many tools serve equally well as weapons, depending on who is wielding them. So it was in the mid-nineteenth century, as controversy over the morality of slave ownership strained the very fabric of the American union and drove the nation into bloody conflict.

Historians agree that the Civil War was a religious conflict as much as it was a political one. Because Americans in both the north and the south regarded Scripture as authoritative, the crisis was essentially a question of interpretation.Footnote ¹⁵ But the nation had yet to develop an institutional infrastructure able to support the robust forms of biblical scholarship that thrived across the Atlantic. Libraries, universities, and professors were sparse at best, even mid-century. Some well-endowed individuals did, however, manage to secure academic training in Germany. Such was the case, for example, with Edward Robinson, an intensely productive orientalist and philologist based at Andover Seminary near Boston, whose influential Bible dictionaries were noted above. Robinson not only studied under Wilhelm Gesenius but went on to publish a translation of his Hebrew lexicon in 1836, the same year in which Robinson’s own New Testament lexicon also went to press (Reference Lee and CarsonLee 2003, 105–115; Reference TurnerTurner 2014, 178–183 and 212–220).Footnote ¹⁶ Few other nineteenth-century Americans had Robinson’s scholarly chops, even if a great many made use of dictionaries like those Robinson produced to understand their Bible. And while the antebellum Protestant context clearly prioritized commonsense literalistic interpretation based directly upon Scripture, in many ways consulting a dictionary was not viewed as a compromise of that hermeneutic. To the contrary, if anything this very theological outlook fostered the production and dissemination of comprehensive reference works as a means of reinforcing itself (Reference Bowman and BrownBowman and Brown 2009, esp. 448–449; Reference GutackerGutacker 2020; Reference NollNoll 2006; Reference OshatzOshatz 2012).

Dictionaries thus stood in the background – and sometimes the foreground – as politicians, ministers, and laypeople alike attempted to discern and argue for the “biblical view” of slavery.Footnote ¹⁷ In so doing, it was not uncommon for interlocutors simply to martial as many Bible verses as possible behind their argument and thus allow Scripture to speak for itself – in English. But at an instinctual level, Protestant interpreters, especially, knew that the meaning of the original languages was paramount in Scriptural interpretation, and for that, biblical lexicons came into view. Of special interest in these debates were the words עבד (ʿebed) and δοῦλος (doulos) – do they mean ‘slave’ or merely refer to some sort of ‘laborer’? It was a crucial question. The patriarchs Abraham and Isaac are each said to have had more than one ʿebed, with no sense of any ethical problems involved (Gen. 20:14; 24:35). Two of the Ten Commandments involve household property, among which the ʿebed is listed (Exod. 20:10, 17). The apostle Paul returns the doulos called Onesimus back to his master, Philemon, again without any censure (Phil. 12–17). Americans grappled mightily with the questions raised by these texts and many others like them.

As Bible readers on both sides of the slavery debate interpreted such passages, the importance and authority of lexicons was often assumed – or affected – without even citing any. For example, in William Reference GrahamGraham’s 1844 pamphlet making a biblical argument against abolition, he cites Hebrew, Greek, and Latin words (in transliteration) and simply declares on the basis of his textual interpretation that “The meaning [of doulos], among the Greeks, was the same with eved [sic] among the Hebrews, a slave” (Reference GrahamGraham 1844, 23). Not long after, on the abolitionist side, Albert Reference 701BarnesBarnes (1846, 64–70) did something similar, noting that “it is of the utmost importance to understand what is the exact sense of the word used to designate this [master/servant] relation,” only to then pronounce the meanings of ʿebed and doulos as authoritatively as a lexicon, but without actually citing one.Footnote ¹⁸ Barnes’ argument was later taken up by A. Pryne in a debate with W. G. Brownlow, who seems to have noted the absence of secondary literature in support of Barnes’ view. Brownlow describes his retort to Pryne in the debate: “the word rendered servants in the Bible invariably means slaves; and I sustained my position by the Bible and the Greek Lexicon” (emphasis added), though ironically he does not say which one (Reference Brownlow1858, 168–169). Whether any reliable dictionary actually supported such proclamations the reader was simply left to presume, a strategy that could only have been effective given the cultural status of the dictionary as a broker of the very historical and linguistic knowledge necessary to understand the Bible aright.

Precisely because of their authoritative status, many involved in the slavery debate did appeal directly to lexicons or Bible dictionaries in their argumentation. For example, in his lengthy pro-slavery treatise, John Henry Hopkins at first merely asserts the original meaning slave for the Greek and Hebrew words, as did the others discussed above. But he then reinforces that assertion in an appendix dealing with the meaning of ʿebed in the Ten Commandments. There Hopkins invokes Edward Robinson’s edition of both Gesenius’ Hebrew lexicon and Calmet’s Bible dictionary, where – in Hopkins’ presentation at least – slave and servant are basically said to have amounted to the same thing in the ancient world. Therefore, for Hopkins the Bible does support slaveholding after all, and he then drives his point home: “The meaning of δοῦλος [doulos], in the New Testament, is hardly susceptible of a cavil. It signifies a slave,” according to several well-known Greek lexicographers that he lists and, “in a word, to all the authorities” (emphasis added) (Reference HopkinsHopkins 1864, 354–355). This is no mere rhetorical strategy. Here in Hopkins’ last comment appears the fallacy ever present to tempt dictionary users. For dictionaries were – and still are – generally considered not just impartial but quasi-scientific, even objective (Reference Lee, Taylor, Lee, Burton and WhitakerLee 2004, 66–67). Thus, with support from such unimpeachable luminaries, Hopkins may safely rest his interpretive case.

Sometimes, however, the appeal to lexicographical authority by pro-slavery advocates came with a conspicuous qualification that speaks to broader social and political dynamics. This qualification surfaces visibly in the pro-slavery writings of the Presbyterian minister Robert L. Dabney. In his discussion of the Abraham narratives in Genesis, Dabney snaps at the arguments of “saucy, meddling, Yankee Abolitionists” and claims that “the word for bondsman [is] […] in every case, ebed,” which he says is “defined by every honest lexicon to mean actual slaves” (emphasis added) (Reference Dabney1867, 106). Here Dabney employs the unusual strategy of casting some doubt upon the authority of lexicons – Yankee lexicons, in particular.Footnote ¹⁹ Elsewhere, Dabney discusses the meaning of doulos and cites the Greek lexicon by “Dr. Edward Robinson, of New York, no friend to slavery,” by which Dabney could only have meant that New York as a state was no friend to slavery, not Robinson, as Hopkins’ citations of Robinson’s dictionaries make clear (Reference DabneyDabney 1867, 148). Similar suspicion appears in a pamphlet by the layman John Richter Jones, who frames his pro-slavery argument as a dialog with certain abolitionist “clerical friends”:

Jones’ mordant sarcasm and condescension is difficult to miss. Yet as inherently untrustworthy as the Yankees were to figures like Jones and Dabney, even dictionaries that issued from northern states nevertheless retained their status as recognized sources of authority for biblical interpretation.Footnote ²⁰

18.3.2 Dictionaries Downstream: Linguistic Theory and Anti-Semitism in Germany

While dictionaries do often serve as tools for fashioning arguments within broader social debates, as explored above, dictionaries have never been just tools. They are treasuries as well; repositories of culturally embedded approaches to knowledge, and therefore never fully impartial. For before there are dictionary users, there must be dictionary makers who (like anyone) are anything but dispassionate bystanders, despite what they may otherwise claim or imagine.

Editorial impact upon dictionaries in biblical interpretation is no better illustrated than by the Theologisches Wörterbuch zum Neuen Testament (TWNT). This multi-author work got underway in 1928 under the direction of the New Testament scholar Gerhard Kittel (1888–1948), who edited volumes one through four of an eventual ten that appeared between 1933 and 1979. By 1964, TWNT began to appear in English as well, translated (but unrevised) by Geoffrey W. Bromiley under the title Theological Dictionary of the New Testament (TDNT), itself complete in nine volumes by 1976 (Reference DankerDanker 1993, 120–122). TDNT was not entirely without precedent (see Figure 18.3). But its exclusive treatment of words judged theologically significant constituted a longer and more confident step in a new direction, one that has earned TW/DNT an unrivaled level of notoriety within biblical scholarship as flawed in both theory and content. These flaws flowed into the dictionary from the cultural and intellectual upstream.

What exactly does it mean for words to be “theological” such that it called for a new genre of dictionary for biblical interpretation? It is a historical question as much as it is a theoretical one, and the answer involves the intellectual impact of Hermes, mentioned above. Harris’ view of language as a manifestation of the thought patterns of a people group was first taken up in the work of fellow Englishmen, such as the philosophical musings of John Horne Reference TookeTooke (1786, Reference Tooke1805), the promotion of Persian studies by the Oxonian Hebraist Benjamin Kennicott (1718–1783), and the diachronic philology-cum-ethnology of William “Oriental” Jones (1746–1794). But Harris’ philosophy of language was more enthusiastically received by scholars in Germany, just as a tide of nationalistic pride was rising along with the academic prioritization of philology as “die Erkenntniss des Erkannten” – the study of the history of human thought and activity, as August Böckh put it in 1809 (see, e.g., Reference HorstmannHorstmann 2010). It was in this dense intellectual milieu of fin-de-siècle Germany that the story of the theological dictionary genre began to unfold, with several protagonists involved.Footnote ²¹ One was the Romantic philosopher Johan Gottfried Herder, who in 1775 applied Harris’ theory to the language of the New Testament (Reference AarsleffAarsleff 1967, esp. 73–114 and 179–182; Reference TurnerTurner 2014, 92–99 and 112–136). To Herder, biblical interpreters needed a dictionary that explained how the thought patterns (Ideen und Ideenreihen) of the Jews, Babylonians, Persians, Egyptians, and others shaped the “broken Greek” of the New Testament. To understand such foreign and mysterious things would then unlock the thought of New Testament writers themselves and, thus, grant access to the divine power and divine wisdom (Göttliche Kraft und Göttliche Weisheit) of Scripture itself (Reference HerderHerder 1775, 15). Friedrich Schleiermacher (1768–1834) also comes into the plotline. In his lectures on hermeneutics, published posthumously, Schleiermacher argued that because the tenets of Christianity as a religious system were first articulated in Greek (apparently picking up some of the idiosyncratic linguistic features of the Septuagint as well), the language itself was changed. The ideas that it conveyed, in other words, made it a new kind of Greek – a Christian Greek. So if biblical philologists were to compile the unique features of this language, that compilation would provide

This very passage was cited by Hermann Cremer (1834–1903) in the preface to his Biblisch-theologisches Wörterbuch der neutestamentlichen Gräcität, a direct forerunner to TWNT that features in Figure 18.3. There, Cremer also affirmed that in the Bible, “the Holy Spirit has been at work moulding for itself a distinctively religious mode of expression […] and transforming the linguistic elements which it found ready to hand, and even the conceptions already existing, into a shape and form appropriate to itself and all its own” (1866, v–vi, quoting Reference RotheRothe 1863, 238). (See also Reference Friedrich and PitkinFriedrich 1976, 638–650 and Reference Voelz, Temporini and HaaseVoelz 1984, 900–910, who suggests [905n69] that Cremer may have misunderstood Schleiermacher.)

Cremer’s view was certainly not without its opponents (Deissmann, most vociferously), but the theory underlying his Wörterbuch – later revised by Cremer’s student Julius Kögel – was the point of departure for Gerhard Kittel’s work in the TWNT. He explains his inspiration in the preface to volume one:

Space precludes elaborating upon the flaws in this understanding of language underlying TWNT. Suffice it to say that this theory has not escaped criticism from scholars in either general linguistics or biblical studies. For development in linguistic theory, see Reference Campbell, Aronoff and Rees-MillerCampbell (2003, esp. 93–100). As to biblical studies, the most penetrating – indeed now classic – critique of the very notion and treatment of “Greek thought” versus “Hebrew thought” in TW/DNT was leveled over fifty years ago by the British biblical scholar James Reference BarrBarr (1961, esp. 1–20 and 206–262). For present purposes, let it be noted that this theoretical flaw shaped TWNT (and the unrevised TDNT) as a dictionary before it ever landed in the hands of its first readers.

But there was more that flowed into TWNT from the intellectual upstream of the German social context in the early twentieth century, and it was far more ominously ideological in nature. In 1933 – the same year that volume one of TWNT went to press – Gerhard Kittel became a member of the Nazi party (Reference 723HeschelHeschel 2008, 184–189). Over the following decade, he lent his scholarly credentials to support the party agenda through research organized under the auspices of the Reichsinstitut für Geschichte des neuen Deutschlands (“Reich Institute for the History of the New Germany”), which itself directly supported Nazi propaganda efforts. Throughout these same years, Kittel oversaw the publication of three more volumes in the Wörterbuch, which he regarded as an exercise in “sacred lexicography.”Footnote ²³

The problem does not stop at Kittel. Numerous contributors to TWNT had similar ideological commitments and political involvement, which itself was often justified in specious theological terms (Reference GerdmarGerdmar 2009, 417–418 and 474–478; and Reference Meeks, Najman and NewmanMeeks 2004). Reference CaseyCasey (1999, 282–286) has identified K. G. Kuhn’s discussion of various words for Jew in the third volume of TWNT (1938) as an example of how this dynamic worked out in Kittel’s dictionary. Although ancient primary sources do not support his claim whatsoever, according to Kuhn – himself a member of the Nazi party – the Greek word Ἰουδαῖος (Ioudaios) is “the non-Jewish” term for the people otherwise linked in genealogical and religious terms. In fact, it “may acquire on the lips of non-Jews a disrespectful and even contemptuous sound” (Reference Kuhn and KittelKuhn 1965, 360). For Kuhn, this occurred because the ancient world had its own serious Judenfrage, which arose after the destruction of the Jerusalem temple in 70 CE and the subsequent rise of “World-Judaism” (Weltjudentum) thanks to the rabbis with their menacingly legalistic Talmud. Kuhn believed that Judaism itself causes Jew-hatred, read this belief into the gaps in ancient primary sources, and published the results in TWNT.

The presence of anti-Jewish and anti-Semitic ideology in dictionaries did not begin with TWNT. The racialization of people groups into biologically framed and hereditary categories had begun in the nineteenth century on diffuse (and often contradictory) theoretical grounds. For example, in the hugely influential, if immodestly named, Grosses vollständiges Universal-Lexicon (“Great Complete Universal Lexicon”), published in 1735, the entry Juden, oder Jüden proclaimed that “Yes, God has marked them in their very natures. Clearly, a Jew has something about him that makes him immediately recognizable and distinguishable from other people. They rouse disgust and horror” (see Reference ManuelManuel 1992, 250–252). To be sure, Kuhn’s entry in TWNT was not as explicitly vitriolic, but there are conceptual parallels. Kuhn’s ideology is rather subtler, interwoven with finer nuances of a scholarly sort that aim to convey impartiality and authority.

Precisely because it is less immediately obvious, the editorial bias in TW/DNT becomes all the more dangerous to the unwitting dictionary user. Still today, TW/DNT commands a status of respectability and scholarly prestige, even among pastors and biblical scholars. Kuhn’s entry is certainly not the only one tainted by similar ideological tendencies and thus useful for or susceptible to anti-Semitic biblical interpretation. Numerous other problematic entries have been identified and discussed by Reference CaseyCasey (1999, 286–291), Reference Nicklas, Lieb, Mason, Roberts and RowlandNicklas (2011, 274–277), and Reference VosVos (1984). As the personal archives of the original contributors to TWNT have been unsealed in greater number in recent years, scholars are now in a far better position to grapple with this issue in a more forthright and informed manner. As recently as 2021, for example, the popular periodical of theological education Didaktikos published a piece by Hans Förster aptly entitled “What to do with the TDNT?” Förster points out that as recently as 2019 TWNT was reprinted in Germany without any revision. After providing two further examples of problematic entries, he answers the question posed in his title: “We should consider every Jewish context issue carefully in the TDNT before we allow ourselves to sit at the feet of a ‘Nazi professor’” (42).

19.3.1 Identity and Representation

Over the centuries, dictionary makers have been motivated by a range of objectives, and among the most significant are those representing sociocultural identity – enregisterment of a society’s lexicon. Given the intertwined nature of verbal expression and sociocultural values, a dictionary inevitably represents a group of people whose members share fundamental aspects of their identity that take form compactly in a dictionary of their language. In fact, still bolder claims than representation have been made. The Swedish lexicographer Bo Reference SvensénSvensén (2009, 1), for example, notes that “It is a commonplace to say that a dictionary is a product of the culture in which it has come into being,” but beyond that, he says, the dictionary “plays an important part in the development of that culture.” (See also Stefan Reference DollingerDollinger 2016, who cites Svensén’s comment.)

19.3.2 National, International, and Regional Pride

Whether they are official dictionaries created by language academies, entrepreneurial ones created by individuals or unofficial institutions, or commercial enterprises produced by great publishing houses, dictionaries have often been motivated by an element of pride – national, regional, or international. The Accademia della Crusca, founded in 1583, and the Académie française, founded in 1635, were expressions of national pride in a language of Italy and France, respectively, and each was initiated with a view to keeping the language pure, which principally meant free from a perceived contagion by foreign words. Such official bodies thus established themselves as guardians. The official guardian of the French language has taken a forbidding approach to lexical borrowings, especially from Anglo-American sources, and in recent times has gone so far as to impose fines for public display of words borrowed from English. Plainly, precluding words from display and from entry in the dictionary is deemed a legitimate tactic for the guardians to keep French and its speakers uncontaminated by Anglo-American words and the values the words embody.

A commission of the Academy recently issued a report that reveals the official view of lexical borrowings from English. Concerning the ninth edition of the Academy’s Dictionnaire, still in progress, its website says this:

The website includes a what-to-say and what-not-to-say section: “« Dire, Ne pas dire », où les Académiciens donnent chaque mois leur sentiment sur les fautes, les ridicules et les tics de langage les plus fréquemment observés dans le français contemporain”!Footnote ² An official focus on native speakers’ frequent “faults” as “laughable” is anathema to the proclaimed objectivity of lexicographical principles and practices elsewhere in Europe – and throughout the world.

While the Academy continues to favor le purisme, many French speakers poke fun at its dictates, and ordinary citizens pay little heed to its pronouncements. Even decades ago (see Reference McDavid, Congleton, Edward Gates and HobarMcDavid 1979, 25), as Professor Guy Forgue of France explained to a group of lexicographers, “the Academy’s dictionary is a joke – except to the Academy itself and to such purists as Etiemble.” That would be René Étiemble, whose book asking Parlez-vous franglais? garnered attention in France through several editions since its initial publication in 1964. Étiemble accused French speakers of polluting their language by importing thousands of words from English, and he particularly lamented American imports. According to French scholar and former Euralex president Henri Béjoint, “Étiemble’s book was one of the most vocal expressions of an attitude that has existed in France since the 18th century, ‘le purisme’, a dedication to the purity of the language, to its protection from foreign influences and to the conservation of its ‘soul’ […], an attitude that can verge dangerously towards bitter nostalgia and xenophobia” (Reference Béjoint2008, 102). Given the social and political upheaval in France and in the United States especially in the 1960s (and after), it is not surprising that the contest between more traditional social, cultural, and political values and more progressive ones continues to play out on the world stage and in the dictionaries that capture the players’ lines. Nostalgia for a world that no longer exists (and in some ways may never have existed) and xenophobia (and the border walls it motivates) are two arenas in which dictionaries, too, contest social ferment and discontent. After all, “every dictionary is a product of its material, cultural and social circumstances, and it can only be properly understood as a result of its historical context and textual tradition,” as Reference Coleman and OgilvieJulie Coleman and Sarah Ogilvie (2009, 2) note. Inevitably, then, by their very nature, dictionaries, like all texts, are not culturally neutral.

Étiemble’s disparagement of French for its lexical borrowings did not go unrebutted. The Dutch lexicographer of French Paul Bogaards disputed Étiemble’s claims. Himself honored by the French government by its naming him “Chevalier dans l’ordre des Palmes Académiques,” Bogaards answered Étiemble’s question even in the title of his book On ne parle pas franglais (Reference Bogaards2008). In the book, he describes a trip to London where, upon arrival at the airport, he spotted a café, a bureau de change, and several boutiques, while in the city he ran across a salade niçoise and a crème brulée in a Prêt-à-manger, bought a baguette and, in a store called “C’est magnifique,” some croissants and brioches. To that list, he adds restaurant, menu, paté, hors d’œuvre, filet mignon, and patisserie, along with many others whose French origins would be lost to most Londoners, the words thoroughly assimilated into the English lexicon. An “incroyable histoire d’amour” is how Bogaards (10) characterizes the relationship between French and English. He was speaking of linguistic facts, but what the Academy and its attitudes display is anything but l’amour.

Like Bogaards, a French international news television network reported in 2020 that “The custodian of the French language […] is routinely criticised for its entrenched conservatism. Its stubborn opposition to any attempt at making French grammar less sexist has left it singularly out of step with society. When it did attempt to feminise a word – ruling that ‘le Covid-19’ should become ‘la’ – it was promptly mocked and ignored” (France 24). There is, then, a chasm between the Academy’s prescriptions and the ways ordinary French speakers use their language. The age-old contest between authority and people power – between guardians and those they would guard – plays out on many stages, including the dictionary.

Differing from the French, officially and unofficially, Italian speakers now borrow words freely, particularly English ones. A possible reason for the difference is that “standard Italian does not stir up linguistic identity for many native users, while English enjoys great prestige as the international language.” That’s according to Carla Reference MarelloMarello (2020, 168), herself an “academic” (as La Crusca’s members are called). Irrespective of its status as a donor language to Italian and French, English has kept company around the globe, stealing countless words (while leaving them put), and English-language dictionaries do not shy away from including borrowings among their headwords when they appear sufficiently in the language corpora that dictionary makers rely on.

Below I’ll report survey evidence showing that, among ordinary citizens, neither the French nor the Italians currently have strong views about their dictionaries as national symbols, but the same isn’t true for other groups at present and hasn’t been so in the past. A dictionary of one’s national language often serves as a mark of national pride, even national identity, precisely because it captures far more than lexical information, even if its semantic and orthographic content are the most frequent prompts for user lookups.

For example, in the preface to his 1828 dictionary half a century after the American colonies declared their independence from Britain, Noah Webster sought to recognize the distinction between the English spoken on American shores and that spoken in Britain during the Georgian era. “It is not only important,” he wrote, “but, in a degree necessary, that the people of this country, should have an American Dictionary of the English Language […]. Language is the expression of ideas; and if the people of one country cannot preserve an identity of ideas, they cannot retain an identity of language.” He pointed especially to differences between British and American governmental structures and practices and the words that denote them, emphasizing the fact that the application of words to social and political structures and events is consequential, as echoed two centuries later in the attitudes of the newspaper opinion writers at the top of this chapter.

Webster honored Samuel Johnson’s belief that “[t]he chief glory of a nation arises from its authors,” saying he did not “expect to add celebrity to the names of Franklin, Washington, Adams, Jay, Madison, Marshall […] and many other Americans distinguished by their writing or by their science,” but, he says, “it is with pride and satisfaction, that I can place them, as authorities, on the same page with those of” such distinguished English writers as Milton, Dryden, and Addison. He also “affirm[s], with truth, that our country has produced some of the best models of composition.” Webster thus expresses pride in his native tongue as spoken and written in the new nation – and, likely, the pride of those fellow Americans who purchased or consulted copies of his massive two volumes. He had high hopes for his work, well beyond meanings and spellings: “I present it to my fellow citizens […] with my ardent wishes for their improvement and their happiness; and for the continued increase of the wealth, the learning, the moral and religious elevation of character, and the glory of my country.” Those wishes extend beyond national pride and national identity, indeed far beyond the lexicon – to learning, glory, happiness, and moral character. Whether entirely realistic or not, such ambitions for a dictionary reveal an extraordinary respect for what many others unthinkingly regard merely as a book of words and definitions.

Scholars have recently offered analysis of other dictionaries that echo aspects of Webster’s about his dictionary. Of the Deutsches Wörterbuch project, which started in 1838 and took 120 years to complete, Volker Reference HarmHarm (2020, 76) notes that it “had a strong national agenda from its very beginning” and showed a history “deeply interwoven with Germany’s long process of becoming a nation and a democracy.” Of the Oxford English Dictionary, first completed in 1928, Michael Reference AdamsAdams (2018a, 38) writes that it was “not just a great historical dictionary but a national dictionary, proof that the English had a language and a dictionary to rival German and the Grimms’ Deutsches Wörterbuch.” The Australian scholar Pam Reference Peters and OgilviePeters (2020, 265) compares the 1981 Macquarie Dictionary in its representing Australian national identity to Reference WebsterWebster’s 1828 Dictionary in representing American identity.

Alongside expression of positive attitudes toward dictionaries as symbols of national pride, a 2017 survey of European dictionary users documents both confirming and disconfirming evidence. From more than 9,500 respondents in nearly thirty European countries, the survey elicited views of dictionaries as a national symbol (among other inquiries). A tentative generalization might appear to indicate that respondents in smaller countries (think Serbia, Georgia, Slovenia) more strongly viewed dictionaries as national symbols, but respondents elsewhere with comparably small populations (think Portugal, Belgium, Czech Republic, and Denmark) did not. By contrast, respondents in a few countries with larger populations – the UK and Germany, for example – did view dictionaries as a national symbol. As the Slovenian lexicographer Iztok Kosem and his fellow surveyors noted, there is no strict parallel between population size and respect for dictionaries as a national symbol; they point to France, Italy, and Spain as countries whose respondents did not highly rate dictionaries as national symbols (Reference KosemKosem et al. 2019, 107). To the extent the French respondents represent a wider French population, this attitude is at variance with that of the Academy though consonant with many of those expressed in popular venues. Indeed, it’s noteworthy that these three countries have had their academies the longest, so something about the disconnect between official and popular sentiment is problematic, and the existence of an official academy charged with guardianship may be that problem.

What is true of dictionaries’ representing national identity for some countries is even more true of their representing regional and provincial identities and the identities of minority groups within larger political and social entities. In a frontmatter section called “the work as a cultural agent” in the Dictionary of Caribbean English Usage (Reference Allsopp and Allsopp1996; DCEU), editor Richard Allsopp writes that it is “[n]o different in this regard from other non-British regional dictionaries when they emerged at landmark times in their nations’ history.” Citing the American Dictionary (Reference Webster1828), the Dictionary of Canadian English (1967), and the Australian National Dictionary (1988), he continues: “the DCEU should be an inward and spiritual operator of regional integration even more powerful as a signal of unity than a national flag would be” (Reference Allsopp and Allsopp1996, xxxi), and he adds that the “weight of evidence supplied in this work should provide sufficient ground to build Caribbean pride to replace the earlier colonial shamefacedness and inhibitions bedeviling this region” (xxxi). Despite criticism for what was viewed as DCEU’s prescriptivism, lexicographer and scholar Lise Reference WinerWiner (1998, 192) acknowledged the “achievement” of the dictionary and exhorted, “May it stimulate much pride […].” Another scholar, writing about Saban English, also a Caribbean language, underscores the importance of regional dictionaries as a source of pride and identity for speakers whose languages have lacked high status: such dictionaries “engage in rewriting the linguistic histories of communities whose language varieties have been labeled ‘broken’ or ‘corrupt’ for centuries. Such linguistic error correction,” Caroline Myrick notes, “can be achieved through the very publication of the dictionary […]” (Reference Myrick2019, 139); such a dictionary “functions as a tool for legitimizing the systematicity and history of the language variety, as well as instilling pride in its speakers” (142); Myrick further notes (citing Reference SchneiderSchneider 2007) that “[i]n areas with histories of colonization, the publication of a dictionary can strengthen national and linguistic identity” (see also Reference SiegelSiegel 2019).

Similar attitudes have been expressed for the English Dialect Dictionary (Reference WrightWright 1898–1905) and the Dictionary of American Regional English (Cassidy and Hall 1985–2013), large-scale multivolume undertakings that reflect pride in dialectal variation within a nation. Likewise for more focused volumes such as the Dictionary of Prince Edward Island English (Reference PrattPratt 1988), the Dictionary of Newfoundland English (Reference Story, Kirwin and WiddowsonStory et al. 1990), Dictionnaire québécois d’aujourd’hui (Boulanger and Rey 1993), and The Liverpool English Dictionary (Reference CrowleyCrowley 2017). Note, then, that for these regional and dialect dictionaries, it isn’t lexical meaning or spelling that’s most salient: it’s representation, identity, pride.

While regional and dialectal representation is a matter of pride, the issue of precisely whose language and identity is represented in a dictionary whose title encompasses speakers of several dialects can be vexing at times and even an impediment to the successful creation of a dictionary. This is perhaps especially true for speakers of an endangered language. Consider Irish, with its three major dialects. In the New English–Irish Dictionary, as editor Pádraig Reference Ó MianáinÓ Mianáin (2019, 161) reports, “when sound files as a guide to a pronunciation were being added to the Irish words online, the decision was taken to systematically provide a sound file in each of the three dialects, as to have chosen one dialect would have incurred the wrath of all those who align themselves with the other two dialects.” Wrath! Not just words and spellings but also pronunciations represent values and identity even within a single nation, albeit one where dialects compete for status. “Ultimately,” Ó Mianáin (167–168) observes, “it is a question of who can claim ownership of modern Irish,” and the same might be said of representation in dictionary projects for some other languages.

19.3.3 Ethnic Identity

Ethnic identity is another human value represented and promoted by the enregisterment of a language variety belonging to speakers whose language may otherwise be viewed as illegitimate or not a language at all. As the Mexican-born American scholar Ilan Stavans writes about the Spanish-language offerings on newsstands in upper Manhattan in New York City in the 1980s, “It was obvious that its authors and editors were americanos with a loose connection to la lengua de Borges.” Regarding the Spanish varieties in those publications, an instructor in the Spanish Department at Columbia University went so far as to claim, “Están contaminaos [sic] […]. Pobrecitos … They’ve lost all sense of verbal propriety,” Stavans reports (Reference Stavans2003, 2).

Stavans quotes Nobel Prize winner Octavio Paz as saying about that variety of Spanish, “ni es bueno ni es malo, sino abominable,” which he translates as “it is neither good nor bad but abominable” and, Stavans adds, “The common perception was that Spanglish was sheer verbal chaos – el habla de los bárbaros” (Reference Stavans2003, 4). He also reports (251) that during a radio discussion in Catalonia that he participated in, another guest (“a language purist affiliated to the Real Academia Espanola de la Lengua Castellana”) claimed that the “mongrel tongue should not be taken seriously until and unless it produced a masterpiece of the caliber of” Don Quixote de La Mancha. Taking up the challenge, Stavans provided an “exercise in literary translation” – a section of Don Quixote – as an appendix to a lexicon of Spanglish that he authored. The exercise serves to rebuke disparaging views of dictionaries of nonstandard languages.

In 2022, Oxford Languages, publisher of the OED and other dictionaries, announced a project to produce the Oxford Dictionary of African American English. The entries would contain senses, pronunciation, and spelling, as well as usage and history. In addition, “each entry will be illustrated by quotations taken from real examples of language in use. This will serve to acknowledge the contributions of African-American writers, thinkers, and artists, as well as everyday African Americans, to the evolution of the US English lexicon and the English lexicon as a whole.” As with several other dictionary websites (Reference Rundell, Hanks and Gilles-MauriceRundell 2017), visitors to ODAAE’s site can take advantage of a section seeking suggestions of words to enter. Contributors are asked about the meaning of any word they suggest and how it is used (ODAAE 2022), and there’s an option to add a sound file to demonstrate pronunciation. ODAAE springs from a sense of ethnic pride and relies in part on web-based crowdsourcing by speakers of African American English. It is yet another example of enregisterment as a hallmark of legitimacy and a source of pride in the language of its speakers and writers – and in the speakers and writers themselves.

19.3.4 Information and Pragmatic Needs

Ordinarily, people turn to dictionaries for specific information, most frequently a word’s meaning: What is an insurrection? But questions about the legitimacy or even existence of an expression may prompt a lookup: Are wanna and supposably words? What about orbisculate? Perhaps spelling (temblor, accommodate) or pronunciation (harass, riposte) perplexes a user. In languages with gendered nouns, are newer ones like Brexit and Covid masculine or feminine? While La Crusca opted for the feminine la Brexit, practice in France and Germany preferred the masculine le Brexit and der Brexit (Reference HenleyHenley 2016). In addition, people often turn to dictionaries to verify their understanding of a somewhat familiar word, as with surreal, marriage, and paparazzi or, as indicated by European survey results (Reference KosemKosem et al. 2019, 103), to learn the meaning of a word they may not have read or heard before or in a specific, baffling context. As an example, on October 12, 2012 the then US vice-presidential candidate Joe Biden used the word malarkey in a debate, and the spike in lookups on Merriam-Webster’s website skyrocketed from the typical few to tens of thousands (see Reference SokolowskiSokolowski 2014).

Aware that users turn to dictionaries chiefly for word meaning, dictionary makers recognize that the senses given in an entry may be complex and challenging to decipher or relatively transparent, depending on the maker’s view of the work and its intended audience. In 1774, John Entick prided himself on the fact that in his New Spelling Dictionary the “Significations are so distinctly comprised and ranged together in one Line, that the Eye takes in at once every thing necessary for understanding, pronouncing, and spelling the Word sought for; which will of course save much Time and Trouble.” As an example, the entry for die, like most other entries, consumes only a single line even in capturing distinct senses: “v. to tinge, taint, expire, lose life.” For the same word, Noah Webster’s Compendious Dictionary (Reference Webster1806) followed a similar pattern but cross-referenced a different entry for information about one sense: “v. to lose life, expire; in the sense of color, see dye.” At quite a remove from the aims of his Compendious, Reference WebsterWebster’s 1828 definition runs an entire column with twelve numbered senses, including “To endure great danger and distress,” as in “I die daily,” the illustrative quotation he cites from St. Paul’s first epistle to the Corinthians. Webster’s dozen senses stand two shy of Johnson’s fourteen, which consume nearly a full column in the 1755 folio, including citations from Shakespeare and the New Testament.

19.3.5 Guidance

Despite the centrality of lexical meaning in motivating users’ consulting dictionaries, they often have other reasons for doing so. Particularly in the public’s eye, one difference among dictionaries is captured in the terms “prescriptive” and “descriptive” (see Chapman, Chapter 16, this volume; Reference Finegan and OgilvieFinegan 2020b; Reference Stamper, Garner and GreeneStamper, Garner, and Greene 2021). While virtually all general-purpose dictionaries are ostensibly – and inevitably – descriptive, they are also inevitably normative – and thus prescriptive. Prescription and description, then, do not represent incompatible opposites. As an illustrative expression of that recognition, the editors of Webster’s New World Dictionary of the American Language (Reference Friend and Guralnik1958) captured the descriptive/prescriptive continuum, contrasting Ambrose Bierce’s view of a dictionary as “a malevolent literary device for cramping the growth of a language and making it hard and inelastic” with their own: “This dictionary was not to create the impression that it was authoritarian, laying down the law about usage; it was to play, rather, the role of a friendly guide, pointing out the safe, well-travelled roads” (Friend and Guralnik 1958, vii). Thirty years later, in its third edition (1988, x), editor David Guralnik boasted that, “[b]y the mid 1970s, the Second College Edition had become the dictionary of first reference for most of the leading newspapers and news agencies in the United States.” But by time the fourth edition appeared (2001 [1999]), the editors were bragging that, “[i]n 1975, the New York Times announced it was replacing the dictionary it had used for decades with the Second College Edition of Webster’s New World Dictionary as its first reference,” and they identified United Press International and the Associated Press as also adopting their dictionary. Then, with lexicographical swagger not unfamiliar in the marketing of some dictionaries in the late twentieth century, they exclaimed that over the years “most leading U.S. newspapers selected the Webster’s New World Dictionary as their dictionary of first choice.” Indeed, the blue and red dustjacket of the fourth edition proclaims it “The Official Dictionary of the Associated Press” – a badge of pride for marketers but of dubious value to most dictionary purchasers.

Not all dictionaries were as sanguine as the 1958 New World. The thoroughly documented lambasting of the 1961 Webster’s Third New International Dictionary for its perceived liberal bent (see Reference FineganFinegan 1980; Reference MortonMorton 1994; Reference SkinnerSkinner 2012; Reference Sledd and EbbittSledd and Ebbitt 1962) helped goad the American Heritage Publishing Company into seizing the opportunity to create what aimed to be a counterpart to Webster’s Third. Seeing the American Heritage company and crew as having “a deep sense of responsibility as custodians of the American tradition in language” and “at a time when the language […] is under constant challenge – from the scientist the bureaucrat, the broadcaster, the innovator of every stripe, even the voyager in space – they undertook to prepare a new dictionary” that would not only “faithfully record our language […] but it would not, like so many others in these permissive times, rest there” (Morris 1969, vi). Thus did American Heritage proclaim itself a linguistic custodian and, with the introduction of a usage panel of judges, distinguish itself from other dictionaries. Subsequently, as Michael Reference 697AdamsAdams (2015) has documented, the dictionary became increasingly descriptive and reached a fifth edition in 2011 before disbanding its staff of lexicographers.

What the American Heritage Dictionary’s publisher and editor tapped into was the fact that “[u]sers don’t want theory, they want practicality: they are seeking the authority of a dictionary, real or perceived, for answers to questions about words,” as competitor Merriam-Webster’s editor-at-large Peter Sokolowski put it half a century later (Reference Sokolowski2018, 129). The Finnish dictionary scholar Juhani Reference NorriNorri (2019, 227) made the same point: “Dictionary users generally expect to be informed about any restrictions on the contexts where a word can be used. Omission or scant provision of such information is likely to lead to criticism.” There is abundant evidence that Sokolowski’s and Norri’s observations are spot on!

19.4.1 Oxford English Dictionary and the Dictionary of American Regional English

The reading program organized by the editors of the OED in the nineteenth and early twentieth centuries is an example of the practice – individuals without specialized lexicographical know-how contribute in myriad ways to the content of the dictionary. For the OED, contributors mainly in the UK and the US took advantage of the nineteenth-century penny post to mail 4x6-inch slips of paper to editor James Murray’s Scriptorium in Oxford, each slip headed by a word and accompanied by an illustrative sentence with identifying information about the source. A few in the crowd submitted tens of thousands of slips, without financial compensation but, apparently, with a more valued reward. The story of the mad surgeon of Crowthorne (depicted in Simon Reference WinchesterWinchester’s 1998 book and a Hollywood film based on it) tells of one extraordinary American contributor who was confined to an asylum in England at the time, but more than 225 are acknowledged by name in the OED’s re-issue in 1933. Among them, one Thomas Austin is credited for 165,000 slips, William Douglas for 136,000, and Dr. T. N. Brushfield for 50,000. While the often overlooked contribution of women to dictionaries is documented by Lindsay Rose Reference RussellRussell (2018), the OED identifies dozens of women who mailed slips to the Scriptorium, including “Miss Jennett Humphreys (of Cricklewood)” for 18,700 slips, “The Misses Edith and E. Perronet Thompson” for 15,000, and “Mrs Toogood (of Kirkby, Yorkshire)” and “Mrs J. A. H. Murray” (Lady Murray) for unspecified numbers. Lexicographer and dictionary historian Reference Ogilvie and SafranSarah Ogilvie (2020; 2023) has documented thousands of contributors over the seventy years it took for the OED to reach Z. That so many readers took enough interest in the project to supply Murray with the millions of citation slips that form the dictionary’s backbone is testament to an admiring and intellectually compelling interest among a wide stripe of readers in that dictionary project and, likely, in dictionaries more generally.

Besides relying on postal systems, as the OED did, dictionaries may be crowdsourced in other ways (Reference Rundell, Hanks and Gilles-MauriceRundell 2017), and the particular attitudes manifest in crowdsourcing activities differ but generally reflect interest and respect. For example, the frontmatter of the Dictionary of American Regional English (DARE, Vol. V) names more than 3,200 contributors of “words and wisdom” to the project, as well as some 1,700 who gave financial support. Such numbers speak volumes about people’s respect for and caring about dictionaries and, in DARE’s case, pride in seeing regional lexicon accurately described (see Reference AdamsAdams 2013, 178–182).

19.4.2 Chamorro–English Dictionary

At the other end of the spectrum from the OED and DARE in “crowd” size is the example of a planned revision of a 1975 Chamorro–English Dictionary. Meetings aiming to engage ten Chamorro speakers in the updating were held in Saipan. Rather than ten, the meetings “revealed that many more community members […] wanted an active role in the revision,” including teachers, librarians, government employees, “and others who felt a strong desire to help their language survive and to take ownership of the process” (Reference 707Chung and RechebeiChung and Rechebei 2014, 309–310). In the end, about a hundred members of the community participated, and the process of meeting to revise the dictionary led to social benefits far beyond lexicography, including increased tolerance between Catholics and non-Catholics (315)!

The importance of the dictionary as a cultural symbol is captured by the fact that, while the original plans also called for a Chamorro grammar and videotaped oral histories, “the community focused exclusively on the revised dictionary as the symbol of a vibrant, full-service language and a central tool for language maintenance and preservation” (310). This then counts as an example of how word meanings and the commitment of those who enregister them are fundamentally about representing people and their sociocultural world in word books. It also exemplifies the sometime soft boundary between the lexicographical priesthood and the lexicographical laity.

19.4.3 Urban Dictionary

One notorious crowdsourced project is Urban Dictionary (UD). Younger contributors especially turn to UD partly to play at lexicography but also to help ensure that the latest slant on words is on record. UD is a noteworthy phenomenon because, in addition to its role as a venue for lexicographical humor, the fact that it is crowdsourced documents a desire among ordinary, younger people to understand words and help others understand them. While UD does not warrant the same kind of respect earned by other dictionaries, even courts of law have turned to it for assistance in understanding informal and slang terms. As a window on how younger speakers used and understood two words in contention, I turned to UD while serving as an expert in a civil lawsuit involving the words sex and gender. UD was one of the most authoritative sources I could identify to help resolve a dispute between a young adult and an older educational institution concerning the meaning of gender on an admissions application form. In a different kind of recognition, the state of Maine has changed the rules about what can appear on vanity automobile license plates and includes UD as one source of information about what ordinary observers would understand a plate to suggest (Reference RevelloRevello 2022). In the judgment of a different crowdsourced dictionary, UD “offers a glimpse of what a community-based dictionary could be, yet the site is an anarchic collection of nonce words and arbitrary example sentences that appear to have been entered solely for the entertainment of the writer and his circle of friends” (Reference Allen, Morrissey and RoeAllen et al. 2011, 132; see also Reference Peckham, Coleman and ColemanPeckham and Coleman 2014).

19.4.4 Dictionnaire Vivant de la Langue Française

That other crowdsourced dictionary is the Dictionnaire vivant de la langue française (DVLF), “designed to give French speakers from around the globe the ability to interact with and share linguistic information for all domains of French usage” (Reference Allen, Morrissey and RoeAllen et al. 2011, 133). Like the website of the Oxford Dictionary of African American English described above, DVLF provides French speakers with an opportunity to contribute words and definitions and otherwise participate in the creation of a rich French lexicon – “to provide more coherent descriptions of emergent usage from Francophone communities around the world. In so doing, DVLF users will move language beyond the restrictions of traditional lexicography, expanding the functionality of the dictionary to an extent rarely before seen, particularly in relation to French dictionaries” (134). In such a project, the transfer of lexicographical clout from guardians to users, from lexicographers to speakers and writers, takes on a more actively involved character by users than that achieved by lexicographers relying solely on even the most up-to-date and representative corpora.

19.5.1 Banning Dictionaries

Despite the respect for dictionaries shown by makers and lay contributors, not everyone finds every dictionary or dictionary entry admirable. As The New York Times reported in 1982, “The American Heritage Dictionary was removed from school libraries and classrooms in Eldon, Mo., because of objectionable definitions it offered for such words as ‘bed,’ ‘tail,’ and ‘nut’” (Reference LandauLandau 2001, 443n15). A later report (Reference ReamerReamer 2022) indicates that the move to suppress the AHD in schools started in Anchorage, Alaska in 1976 and spread to cities in Indiana and, later, California and Nevada. (It would be unfair not to note some strong opposition to the Anchorage ban, with one Alaska state legislator calling it “absolutely ludicrous,” another referring generally to a dictionary as “a most sacred document,” and a third, perhaps hyperbolically, predicting about the suppressors, “Now I suppose they’d like to go for the Bible. Lots of good words with dirty meanings there” (Reference ReamerReamer 2022). In 2010 a school district in California withdrew the tenth edition of “Merriam Webster’s Dictionary” from classrooms because it included a definition of “oral sex” (Reference FloodFlood 2010 notes it isn’t clear just which dictionary that was). More recently, dictionaries of Spanish and Swahili have been banned in certain US prisons for fear inmates communicating in those languages would leave guards in the dark about prisoners’ discussions (NPR 2022).

19.5.2 Beatifying Dictionaries

By contrast, precisely because of how they treat certain kinds of lexical information, including definitions and illustrative quotations, dictionaries can be kept in print and endorsed even centuries after first publication. Almost 150 years after Webster’s American Dictionary appeared, it was reprinted in 1967 in an inexpensive facsimile edition – not for historical reasons but for its moral and political content. In the frontmatter of the reprint, the Foundation for American Christian Education includes an essay about Webster’s political and religious beliefs and an apologia that stresses the values of morality and representative government. “Today the field of lexicography has been demoralized by those who would make ‘contemporary usage’ and ‘slang’ a standard of reference for students in our schools. […] We need the ‘primary’ Biblical, Christian and Constitutional meanings of words,” said the introduction’s writer (Reference AnonAnon. 1967, 10). Captions in the apologia such as “Biblical Definitions Restored” and “Christian Constitutional Meanings Restored” (9–10) are telltale. But most revealing about the continuing importance of Webster’s dictionary for some modern-day Americans are statements like this one:

Noah Webster was not only a patriot: he was also a born-again Christian, and his dictionary relied heavily on quotations from the Bible to illustrate words in use. Thus has it become a hallowed reference work. (For more on Webster and religion, see Reference FineganFinegan 2020a.)

19.5.3 Protesting Individual Entries

More frequent than dictionary banning or beatification are protests against particular dictionary entries or particular senses of words by activists who want (potentially) offensive words and senses to be accurately described or in some cases elided altogether from a dictionary. There are “moments when dictionaries are perceived as anything but delightful, comforting, wholesome, uniting, and instead seen to disturb, disparage, denigrate, and divide people,” Lindsay Rose Reference RussellRussell (2021, 236) has noted. There is also self-censorship, when editors or publishers choose to omit certain headwords or senses. Dictionary editors and subeditors select their illustrative quotations from what is often a plethora of possibilities, and self-censorship may affect those choices.

19.5.4 Lexicographical Self-Monitoring

In the frontmatter of the Dictionary of Southern Appalachian English, the authors write that, while they had always intended “to base this work on language as it has historically been used in Appalachia and beyond, there are inevitably terms and ideas in the samples illustrating head words […] that do not reflect the authors’ feelings, attitudes, or opinions. Although it is the obligation of the researcher to present information in a truthful way,” they concede, “there are ideologies that need not be perpetuated in text. As such, material of this nature has been largely removed from this volume. […] [I]t is important to remember that, as modern readers, the lens through which we view historical language is vastly different from that of the speakers whose words are presented herein” (Reference Montgomery, Jennifer and HeinmillerMontgomery and Heinmiller 2021, Authors’ Note). The authors do not identify any omitted words, but it’s easy to imagine candidate citations containing words easily replaceable by equally good citations judged to be lacking in offense.

In another context, the distinguished lexicographer Sidney I. Landau acknowledged that he had often found himself “unable to use an otherwise excellent corpus citation to illustrate a perfectly inoffensive term because it was embedded in a context that included disparaging or grossly insensitive comments about women or a minority group. To include such comments even within quotation marks would invite charges of bias and insensitivity from readers.” He adds this: “In determining whether a particular usage is insulting or not, the lexicographer is compelled to use his own experience, moderated of necessity by his own moral views, whether consciously or not” (Reference Landau2001, 234). While the selection of quotations for some online dictionaries is done automatically, print dictionaries and, at present, probably most online dictionaries rely on lexicographers to select appropriate quotations to illustrate usage. As mentioned above, Webster relied heavily on quotations from the Bible, so much so that his 200-year-old dictionary is in print today in the service of certain religious (and civic) convictions.

19.5.5 Ethnic Sensitivities

In the US, use of the so-called “N” word offers such offense in many or most situations that dictionaries entering it as a headword, and the entire Merriam-Webster enterprise, have been threatened with boycott (Reference BaniskyBanisky 1997, 1). Even entirely unrelated words mistakenly interpreted by some as related to the “N” word have drawn protest. Consequently, in some dictionaries, the unrelated words display a usage note reporting that fact.Footnote ³ As Geoffrey Reference NunbergNunberg (2001, 96) has astutely observed about such misunderstood words as niggardly and picnic whose sounds suggest quite different but sensitive words: “Phonetics always trumps etymology! ”

Other ethnic terms (including spic, dago, kike, gypsy) have been targets of objection. Robert Burchfield, editor of the four-volume Supplement (1972–1986) to the OED, says it wasn’t clear when “the battle cry” against dictionaries first arose, but he cites a 1920s dispute when the Jewish Chronicle expressed “no small gratification” that the Clarendon Press, in response to a protest, labeled the “sinister meaning” of Jew derogatory. Protest against this single word in the Pocket Oxford Dictionary gave rise to similar objections to certain senses of Turk, Tartar, and Jesuit, he reports (Reference Burchfield, Michaels and Ricks1980, 15–16). By contrast, as recently as 2022, the Associated Press reported that “[t]he leading dictionary of standard German has changed its definition of Jew, or ‘Jude’ in German, after a recent update caused an uproar in the country’s Jewish community […].” Duden had explained that “occasionally, the term Jew is perceived as discriminatory because of the memory of the National Socialist use of language,” and it recommended that “formulations such as Jewish people, Jewish fellow citizens or people of the Jewish faith are usually chosen.” According to the AP, “The explanation led to an outcry from leading Jewish groups and individuals who stressed that identifying themselves or being called Jews is not discriminatory, in contrast to what Duden’s definition implied” (Reference GrieshaberGrieshaber 2022).

19.5.6 Socioeconomic and Class Sensitivities

In a decision related not to ethnicity but likely to socioeconomic and class status, the Oxford Advanced Learner’s Dictionary, at the instigation of a campaign by a group calling itself the Essex Girls Liberation Front, made the socially sensitive decision to excise an entry that defined Essex girl as “a name used especially in jokes to refer to a type of young woman who is not intelligent, dresses badly, talks in a loud and ugly way and is very willing to have sex” (Reference MarshMarsh 2020). From such examples, then, it is clear that when dictionaries define words that characterize ethnic, religious, and less established or organized groups, they risk treading on flammable sensibilities.

19.5.7 Other Taboo Words and Senses

This chapter lacks space to address attitudes toward dictionary treatment of the notorious four-letter words and others viewed as obscene or vulgar, but those words are treated by Reference LandauLandau (2001, 228–231; see also Adams, Chapter 30, this volume; Reference FineganFinegan 2003). Still, it is sobering to highlight two kinds of omission – one involving Victorian-era sensibilities about sex, the other involving mid-twentieth-century American language guardianship. First I should point out that to illustrate the sense ‘to experience a sexual orgasm’ for the verb die, OED2, the late twentieth-century edition, cites Claudio’s “I know who loues him and in despight of all, dies for him” (from Shakespeare’s Much Ado About Nothing).Footnote ⁴ The first edition likewise cited Claudio’s line to illustrate the same verb. But in Victorian times “dies for him” meant only ‘to suffer pains identified with those of death; to languish, pine away with passion; to be consumed with longing desire.’ (No sex, please, we’re British! as the 1970’s dramatic farce put it.)

The other example comes from the American Heritage Dictionary and involves an entry unlikely to be regarded as sexy. The 1969 AHD, published (as noted earlier) in reaction to the perceived liberal bent of Webster’s Third, enters two senses for unique: ‘solitary’ and ‘unparalleled’. But it also adds a usage note that assesses a third sense (which later editions of the dictionary define as ‘unusual, extraordinary’). Just as OED1 had citations pointing to the sexy sense of die, AHD had evidence of the ‘unusual, extraordinary’ sense of unique; indeed, the first edition included a usage note reporting that 94 percent of its usage panel judged that sense “unacceptable.” As a consequence, anyone who consulted that first AHD would be only obliquely aware, if at all, of the third and doubtless very frequent sense. Arising from different spheres of sensibility, Victorian and mid-twentieth-century language guardians chose to protect users, with dictionary makers in both instances feeling free to exercise their judgment of good and bad in language. Today, at least in the US, as noted above, entire dictionaries are being cast out of schools and school libraries because they include definitions deemed too graphic for children’s minds – the very definitions, in many cases, that eager teens and tweens are keen to spy. Such adolescent users might well wonder what good a dictionary is if it doesn’t define the very words they’ve sought to look up.