Effable characters: the problem of language and its media in seventeenth-century linguistic thought

ABSTRACT

What did ‘language’ mean in seventeenth-century Britain? Was it prototypically oral, as most modern-day linguists will allow, or was ‘language’ understood to encompass written and spoken forms in equal measure? This article launches those questions with an eye to the philosophical language schemes of John Wilkins (1614–1672) and George Dalgarno (c. 1616–1687). It challenges the assumption that early modern thinkers’ conceptions of ‘language’ referred exclusively to the oral and argues instead that both Wilkins, Dalgarno, and others in their orbit understood ‘language’ to encompass both oral and visual communication in a non-hierarchical way. Designing a philosophical language meant attending to both its written and its articulated forms, and both Wilkins and Dalgarno defended the potential of each medium to communicate thought directly, while grappling with the mundane problems that each one posed.

KEYWORDS:

• Early modern
• language
• real character
• philosophical language
• spoken and written language
• John Wilkins
• George Dalgarno

Acknowledgement

My thanks to Ann Blair, David Cram, Michael Gordin, Tony Grafton, Sean O’Neil, and Leah Whittington. Earlier drafts of this paper were workshopped at the Pomerium in the English Department at Yale and the Early Sciences Working Group in the History of Science Department at Harvard, with responses by Caitlin Hubbard and Austen Van Burns, respectively, and I owe enormous thanks to both and to all the workshop participants. Special thanks, too, to the editors, contributors, and reviewers of this special issue, who have provided exceptional support and feedback throughout.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ For succinct and up-to-the-minute introductions to the issue, see Cram and Neis (2018, 1–2) and Gorman and Sproat (2022). Classic critics of the paradoxical prioritisation of speech over writing are Harris (1980 and his following oeuvre) and Linell (2005 and his work before and since), both of whom assert that modern linguists are inheritors of a longer, intellectual legacy that, since Aristotle, has recognised writing as little more than a transcription of speech. More direct intellectual origins for the present prejudice against writing are Saussure ([1916] 1983), Bloomfield (1933), and Chomsky (1957), all influential works of the past century that position speech as a linguist’s primary object of study. That position has raised the hackles of certain contemporaries, who have argued that writing and speech are interdependent media (Abercrombie [1937] 1965; Bradley 1913; McIntosh 1956, [1961] 1966; Uldall [1944] 1966; Vachek [1945] 1966, 1973) and, more recently, that writing is a tool of inherent linguistic analysis (Daniels 2013; Davidson 2019). Those linguists who have taken writing as their object of study (beginning with Gelb [1952] 1969 and most recently encapsulated by Meletis 2020) can attest to ongoing disciplinary prejudices (Meletis 2021). Within the spheres of philosophy, literary criticism, and cultural studies, Derrida ([1967] 2016 drew vocabulary from Gelb to subvert the secondary place of writing granted by Saussure, his structuralist successors, and his Enlightenment predecessors (namely Rousseau), but Derrida’s De la Grammatologie is not particularly well-regarded by linguists or linguistic historiographers. Harris, for example, concluded that Derrida’s interpretation of Saussure ‘is academically worthless’ (2001, 188), and while many of his reviewers call Harris’s judgement uncharitable, they do not rush to Derrida’s aid (Matthews 2001; Sanders 2004).

² Oxford English Dictionary, ‘Language’; Dobson ([1957] 1968, §421 n. 7, p. 983).

³ I have confirmed this claim against various searches for ‘language’, ‘lingua’, and ‘langage’ on the Lexicons of Early Modern English database. For direct comparanda, see, Florio (1598, 205; 1611, 286); Cawdrey (1609, sig. F3^v; 1613, sig. F3^r; 1617, sig. F3^r); and Cotgrave (1611, sig. Bbbiv^r). Bullokar (1616, 1621, 1641) and Phillips (1658) do not include the term.

⁴ For a short and whimsical warning against reading Cawdrey with too much earnestness, see Gleick’s review of the edition of Cawdrey’s Table Alphabetical published by the Bodleian in 2007. For a less snarky account, see Simpson’s introduction to that very edition (esp. 21–7).

⁵ Fleming (2017) and Maat (2004) are particular advocates of this view and their arguments are discussed in greater detail below.

⁶ The word ‘linguist’ has been in use since the 1580s, but until the early nineteenth century it described a skilled learner or practitioner of language, not a scholar as such. Hence the opaque definition of linguist in the dictionary companion to Wilkins’s Essay: ‘adj. Language…artist’ (sig. Kkk1^v). As a result, there are both etymological and disciplinary reasons to shy away from using the terms ‘linguist’ and ‘linguistic’ of pre-modern thinkers (the latter outlined by Lewis 2007, 9). Even so, expunging the modern valences of ‘linguist’ and ‘linguistic’ in histories of pre-modern thought imply that those periods lack rigour or sophistication. Just as there is power in understanding the work of natural philosophers as ‘scientific’, there is power in referring to the work of language planners as ‘linguistic’, and I will do so (carefully) throughout this article.

⁷ In contrast to the previous note, seventeenth-century language planners regularly employed the terms ‘artificial’ and ‘a priori’ to describe their inventions. ‘Artificial’ (as in, designed by human artifice) was used to contrast the invented language schemes with ‘natural’ languages, while a priori denoted a system whose vocabulary was derived from a ‘rational’, ‘reason’-based study of nature, rather than borrowings from pre-existing natural languages (this was, at least, the aim). For a succinct, theoretical-cum-historical discussion of a priori versus a posteriori language construction, see Okrent’s annotated bibliography of constructed languages (2013).

⁸ This is not to disparage dictionaries or grammar books, merely to point out that their intentions are practical and their contents less prone to theoretical discussions of language, media, and the mind. On grammar- and language-teaching in early modern England, see: Schrire (2021, 2022) (on the material and cognitive aspects of grammar-teaching); Gallagher (2019) and Michael (1987) (on language instruction); Michael (1970) and Vorlat (1963) (for the development of grammatical categories); and Robins (1993); Gwosdek (2000, 2013); and Alston (1974) (for context and bibliography).

⁹ For scholarly treatments of Lodwick, see Henderson and Poole (2009, 2011) as well as Salmon (1972). Most discussions of Beck appear as an ancillary to larger treatments on the language planners, but see Salmon [1976] 1988 for a dedicated account.

¹⁰ The minute books of the Royal Society record that on 29 October 1662, ‘D^r Wilkins was put in mind to prosecute his universall language’ (100); that on 13 April 1668 ‘The Council licenced D^r Wilkins his Book, entitled An Essay towards a Real Character and Philosophical Language’ (159); and that on 26 July 1699, ‘S^r John Hopkins proposed y^t D^r Wilkins his Universal Character translated into Latin by M^r Ray, now in D^r Hook’s hands, should be printed’ (141). There is ample correspondence between and beyond members of the Royal Society relating to Wilkins’s and Dalgarno’s language schemes, best accounted for in Lewis (2007) and most comprehensively available in the edited correspondence of Henry Oldenburg (1965–1975) and digitised papers of Samuel Hartlib. For collected and edited correspondence relating to Dalgarno’s work, see Cram and Maat (2001, 415–430).

On the reception of Wilkins’s Essay and the erudition of its target audience, see Lewis (2002, 139 and 145 note 28). The wide circulation of Wilkins’s Essay is apparent from the sheer number of surviving copies: 163, according to the English Short Title Catalogue. The physical and intellectual profile of Wilkins’s Essay, a 600-plus-page folio published by the Royal Society and adorned with wide margins and fine engravings throughout, gives it a much greater survival bias than Dalgarno’s slim, self-published octavo, so while the twenty-one extant copies of Dalgarno’s Ars signorum may, in contrast, seem rather paltry, it is nothing to be sniffed at. It can be presumed (and, in Francis Lodwick’s case, proven) that the language planners each owned copies of their competitor’s works (Henderson & Poole 2009), but the books circulated further afield, too. On 30 August 1681, John Locke sent copies of both Dalgarno’s Ars signorum and his recently-released Didascalocophus (1680) from London to Nicolas Toinard in Paris, along with the recommendation that Toinard consult Wilkins’s Essay if he could read English. Ars signorum was by then twenty years off the press, and Locke could find only the one copy ‘in all the bookstores in London or Oxford’ (‘en touts les boutiques libraires à Londres ou à Oxford’).

For examples of users’ annotations and engagement in Dalgarno’s Ars signorum, see the copy previously owned by John Wallis, who added an interlinear translation of the preface (written in Dalgarno’s proposed language) into English (Oxford, Bodleian Library, Savile Cc 18) and the copy owned by John Byrom in the eighteenth century but heavily annotated in a seventeenth-century hand (Manchester, Chetham’s Library, 3.F.4.2); for his Didascalocophus, see the lightly-annotated copy likely owned by the English Hebraist Samuel Barker (1686–1759) (Manchester, John Rylands Library, R172476). Wallis’s edition of Wilkins’s Essay bears similar signs of use (Oxford, Bodleian Library, Savile A 4), though the most heavily-annotated copies of the Essay I have come across remain those at the John Rylands Library in Manchester (SC3311D) and at the Newberry Library in Chicago (Bonaparte 11544). Fleming offers a thorough description of the eleven copies he consulted, and the annotations therein, in his discussion of how the Essay was read (2017, 214–217).

Finally, on popular satire of the language planners, see Swift’s farcical description of the academics of Lagado in Gulliver’s Travels, who sought to do away with words entirely by having everyone carry around ‘Bundles of Things’, so that speakers might converse by brandishing about concrete objects, thereby avoiding any lexical ambiguity (1726, II.3.5, 75–79). For more on the intersection of constructed languages and literature, see Knowlson 1975.

¹¹ Here I cite this very journal issue, in particular the contribution by Sean O’Neil. For a succinct and multilingual list of classic scholarship devoted to the language planners, see Lewis (2007), 2–4.

¹² See, for example, Wilkins’s ‘natural character’ (1668a), Lodwick’s ‘perfect universall alfabeth’ (1658, 1686 and two undated manuscripts), and William Holder’s ‘natural alphabet’ (1668a-b, 1669). On the idea that a single alphabet could be used to phonetically render any language, a notion which in the early modern period was quite new, see Cram (2018).

¹³ For a classic account of these manifold linguistic projects and their mutual motivations, see Salmon [1966] 1988.

¹⁴ The criticism most commonly associated with Beck’s invention was levelled by Dalgarno, who described Beck’s invention as ‘nothing else, but an Enigmaticall waye of writing y^e English Language’ which ought not to have been published. Dalgarno’s ‘impartial Judgment’ (or, by his own label, ‘rash censure’) (Cram & Maat 2001, 417–418) appears not to have been shared by all. Nathaniel Fairfax, writing to Henry Oldenburg on 29 January 1667/8, described Beck much more positively, as ‘a Divine of steddy reasonings, shrewd reaches, narrow searchings, [and] Mathematically given’ (Oldenburg 1965–1975, vol. 4, 123–126 and 335–338, here 125–126). Beck himself pitched his own invention as a ‘Mechanical’ alternative to Wilkins’s ‘incomparably performed’ scheme, more suited to the ‘Capacities of ye Vulgar’, for whom ‘natural Logick or Metaphysick’ was not a regular object of study. In a letter to Henry Oldenburg on 15 August 1668, he suggested that the two schemes coexist, and added that it would be ‘no hindrance to yt Philosophical Character nor no burden to ye world if mine were recommended as a Pocket Mercury to Travaylors’ (Oldenburg 1965–1975, vol. 4, 16–17). A month earlier, on 17 July 1668, Nathaniel Fairfax had attested to Beck’s more humble intentions in another letter to Henry Oldenburg. On Wilkins as the ‘culmination’ of the language schemes, see Salmon [1974] 1988.

¹⁵ Maat’s first statements to this effect are gingerly addressed in his collaborations with Cram, in which both authors urge for a distinction and consideration of the relationship between writing and speech (1998; 2001, esp. 4–6). Maat’s single-authored monograph, Philosophical Languages in the Seventeenth Century (2004, developed out of his 1999 dissertation), takes a stronger stance (see 16, 19, 21, 42, and 44).

¹⁶ For a clear treatment of the convoluted meaning of ‘letters’, see Abercrombie [1949] 1965.

¹⁷ Discussions of Bacon’s ideas of language and their influence on the language planners are to be found in practically all histories of the subject, but for a particularly robust discussion see Lewis (2007), 6–20.

¹⁸ ’‘Eστι μὲν οὖν τὰ ἐν τῇ φωνῇ τῶν ἐν τῇ ψυχῇ παθημάτων σύμβολα, καì τὰ γραφόμενα τῶν ἐν τῇ φωνῇ’. For a less ambiguous translation, consider that of Ackrill (1984): ‘spoken sounds are symbols of affections in the soul, and written marks symbols of spoken sounds’ (§1).

¹⁹ As an aside emblematic of the interconnectedness of Wilkins’s work with wider, language-related publications, it is worth pointing out that in his discussion of musical language Wilkins cited ‘Nuntius Inanimat’, the tonal language sung by the lunar residents in Francis Godwin’s The man in the moone (1638), produced by the same printer as Wilkins’s Mercury.

²⁰ See Dalgarno’s discussion of haptics (‘haptology’) in Didascalocophus (pp. 31, 87–90).

²¹ In his Cryptomenysis patefacta (1685), Falconer’s overarching depiction of language echoed Wilkins’s framework at every turn: ‘The way of Communicating our Thoughts’, Falconer wrote, ‘is reckoned threefold; or, there are three means of Converse, viz. By SPEAKING, by WRITING, or by SIGNS and GESTURES. Words are, or ought to be the Picture of Conceptions, as Letters are the Images of Words, the Tongue speaks to the Ear, but the Gesture to the Eye’ (118). While greatly indebted to Wilkins’s Mercury, Falconer’s framework ought not to be read merely as obsequious parroting of his sources, but as an espousal of the same linguistic construct. Elsewhere in his book, Falconer hotly disavowed the language planners’ efforts, suggesting that they concentrate their energy on reconstructing the Adamic language or else – if the goal was secret communication – learn ‘some of the Eastern Languages’ or retire into a corner and converse in whispers (126–127).

When it came to the relationship between codes and language, Falconer was quite firm on the fact that decipherment and translation were two separate processes, and that to conflate them was a ‘Fallacy’ (sig. A7^v). He maintained that translation is an art, decryption an equation, and the former cannot be achieved by merely following the algorithmic instructions on a key (118–124). Deciphering unknown languages may give a person motivation to learn those languages – ‘a desire of understanding the meaning of what is so Decyphered, and Desire (joyned even with a Superficial knowledge) a Facility of attaining it’ – but no more (sig. A7^v). Recent scholarship has emphasised that most early modern cryptographic manuals in England and the Continent did more than treat the alphabet as a platform for limitless permutation, and that they also encouraged and revelled in multimodal creativity, personalisation, and language play (Ellison 2014; Grafton 2023, 74–79, 141–147). But the conceptual boundary between language and encryption (variously and ambiguously referred to as ‘code’, ‘cipher’, or ‘steganography’ by writers in the period) appears firm.

²² The full classification scheme is as follows.

Genus IX, ‘the Predicament of Action’

> Difference II, ‘CORPOREAL ACTION’

> Species III, ‘The Corporeal ACTIONS peculiar to Men, or the several wayes of expressing their mental Conceptions’,

‘Articulate sounds’, and

‘Figures representing either words or things’.

It is worth noting that the entry for ‘SPEAKING’ contains twenty-six alternative terms, which together emphasise the orality of speech far more than its linguistic nature. The terms, in full, are: ‘talk, utter-ance, mentioning, Elocution, pronounce, express, deliver, Prolation, Spokes-man, effable, voluble, fluent, say, tell, mutter, mumble, jabber, jargon, vein, Grammar, Rhetoric, Oratory, Eloquence, Prolocutor, nuncupative, by word of mouth’ (235).

²³ Proper nouns posed a conundrum for many of the language planners. Beck (1657a) excused them from his universal character entirely and allowed them to ‘be writ in the usual letters’ on the basis that they ‘sound and signifie very like to all Nations’ (10, see too 34). Lodwick took a creative approach in his Common writing, suggesting first that a real character be constructed from the proper noun’s etymological root, second that it be based on the character or quality of the proper noun, and third (should all other options prove ‘too tedious’) that the proper noun should simply be written in the ‘ordinary character’, as in Beck’s system (1647, 10).

²⁴ My thanks to David Cram for the details of Dalgarno’s biography beyond those presently published in the Oxford Dictionary of National Biography.

²⁵ In a letter from Faustus Morstyn to Samuel Hartlib on 10 April 1657, Morstyn describes Dalgarmo’s adequate position in society (‘Est natione Scotus, genere nec Nobili, nec prorsus ignobili ex generosis enim (ut asserit) parentibus satus’) and paints a romanticised view of Dalgarno’s pecuniary position: ‘He is of a virile age, is in somewhat strained circumstances, and has a lively intellect whose acuteness has been whetted by poverty, that sister of a good mind’ (‘ætatis est Virilis, sortis tenuis, vegeti ingenij, cuius vires maximè acuit bonæ mentis soror paupertas’).

²⁶ In the spirit of fundraising, Dalgarno convinced Richard Love, Master of Corpus Christi College, Cambridge, to sign his name to a broadside testifying to the dexterity, rationality, and modesty of Dalgarno and his ‘artificial new language’, ending with a pointed comment that it would be ‘a thousand pitties’ should Dalgarno not be able to complete his project ‘for want of help at home’ (‘Broadsheet 4’ in Cram and Maat 2001, 113–116, here 115).

²⁷ ‘Broadsheet 4’ and ‘Broadsheet 5’ in Cram and Maat (2001, 113–135, quotation on 116). The number of patrons declines across the three lists: in 1658 (‘Broadsheet 4’, 116), Dalgarno lists fifty names in print and manuscript; in 1660 (‘Broadsheet 5’, 118), there are thirty-four; in Ars signorum (1661, sig. A8^v), there are thirty-one. Twenty-seven of the names appear on all three lists, twenty-one appear on only one, and eight appear on two.

²⁸ There were other, less generous donors, Dalgarno tells us later, but due to ‘the restricted space available in the present booklet’, their thanks had to await another venue (‘Broadsheet 5’, 135).

²⁹ Dalgarno to Hartlib, 20 April 1657; to Pell, 26 December 1657; to Brierton, 17 February 1658; to Hartlib, 20 May 1658, and 3 November 1659. All are edited in Cram and Maat (2001), 417–427, quotation on 417.

³⁰ Cram and Maat, ‘Broadsheet 3’, 2001, 111.

³¹ Cram and Maat, ‘The Autobiographical Treatise’, 2001, 358. By Dalgarno’s account: ‘All my former and great hopes and expectations for a reward of my labours were almost quite defeated. Both Scholars and benefactors fell off; some civilly excusing themselves that they would come in, some to learn, others to put to their encouraging hand, when they should see the deseigne established by the approbation of learned men’ (358–359). When John Owen, the person who had first introduced Dalgarno to Seth Ward and thence to Wilkins (356), published a brief account of the universal character in his defence of the Bible, he made no mention of Dalgarno by name (apparently believing Dalgarno to have abandoned his efforts) and credited only Wilkins as having brought the language ‘to perfection’ (1659, 277), a citation slight that Dalgarno noted with ire in his autobiography (359).

³² Unless otherwise noted, translations of Dalgarno are from Cram and Maat 2001.

³³ For all that Dalgarno maintained that he and Wilkins remained friends, Dalgarno had little trust in his former colleague. In a letter on 3 November 1659, Dalgarno asked Hartlib to keep him abreast of any developments in the schemes of ‘Comenius, Sir Tho. Urquhart my Contrie man, or Dr Wilkins, who, if he publish anie thing he deales neither ingeniously nor justly with me & I feare if he attempt anie thing on this subject he shal have small credit of it’ (Cram and Maat 2001, 426–7).

³⁴ Cram and Maat, ‘Broadsheet 2’, 2001, 90. See too Dalgarno’s treatment of the subject in Ars signorum: ‘As knowledge is transferred from one person to another by the instrument of the ear by means of sounds, so it is also by the instrument of the eye by means of figures’ (‘QEmadmodum scientia transfertur a homine ad hominem per organum auris, mediantibus sonis; sic etiam per oculi organum mediantibus figuris’) (1661, 12).

³⁵ ‘hæ enim nihil sunt aliud qam Signa Signorum, id est sonorum; adeoqe illis necessario posteriora’ (1661, 2). For more on Dalgarno’s attitudes towards Aristotle, see his commentary ‘On Interpretation’ (Cram and Maat 2001, 391–408).

³⁶ ‘nam non minus certum videtur usum Characterum, qatenus Rerum ipsarum & mentis Conceptuum immediate significativi sunt, ante hanc prius longe cognitum fuisse’ (1661, 2).

³⁷ Cram and Maat, ‘Broadsheet 2’, 2001, 90.

³⁸ ‘Hic diligenter est advertendum qod eadem sit Ars signorum audibilium & visibilium’ (1661, 13).

³⁹ ‘est qod non nimus naturale sit homini communicare in Figuris qam Sonis: qorum utrumqe dico homini naturale; licet Scribere has vel illas figuras, vel loqi has vel illas voces, sit omnino ad placitum’ (1661, 2).

⁴⁰ This juxtaposition of ‘this Dumb Character’ and ‘Language’ would seem to support Maat’s argument for their distinction, were it not for the fact that in the same manuscript and elsewhere, Dalgarno described his dumb character as a ‘Language’ in its own right. While discussing his initial plans for his character in his autobiographical treatise, Dalgarno recounted how ‘the body of Language, by which I understand the commone notions of familiar and ordinary intercourse and dayly use in vita communi, secluding terms of Art and the numerous species of Natural bodies, might be exhibited both Dictionary and Grammar upon one face of a sheet of paper’ (355). Earlier, in his 1658 broadsheet, ‘News to the whole World’, Dalgarno testified that ‘The Art hereof, shall not only rest onely in a dumb Character, but by the same Rules it shall be made effable, in distinct and dearticulate sounds’ (2001, 109). In other words, Dalgarno acknowledges that his ‘Dumb Character’ was never, in fact, ‘Dumb’, and could be pronounced from the outset.

⁴¹ ‘pauci apprehendant Artem Sonorum & Figurarum esse omnino eandem’ (1661, 14).

⁴² ‘Cum tamen nemo Artem communicandi per figuras tradere potest, qin eadem Characteris præcepta sint eadem ratione sonis applicabilia’ (1661, 13). See too his later point on p. 15: ‘whatever can be expressed in figures can by the same method equally well be expressed in sounds’ (‘qicqid præstari potest in figuris, idem eadem Arte præstari potest in sonis’).

⁴³ ‘An non qi legit hunc librum, habet me ipsi communicantem animi mei sensa in Charactere Muto; non enim audit meas voces? Sed dicet, hic Character est effabilis’ (1661, 14). See also Dalgarno’s discussions of the literacy of the deaf in his autobiographical treatise (2001, 381) and Didascalocophus (all, but esp. 1–2, 7, 52).

⁴⁴ As Dalgarno put it in his autobiographical treatise: a real character ‘layes a double burden upon the memory by obliging it to learn two distinct words for one thing’ (Cram & Maat 2001, 363). In 1677, Robert Plot remembered this shift from ‘dumb Character’ to ‘a Language’ as the crux of Dalgarno’s innovation, agreeing that the resulting system had ‘this advantage over any Character, that we may use our known Alphabet to express it, whereas in a Character the figures must be new’ (283).

⁴⁵ ‘Characteres prius noti omnibus Europæ populis’ (1661, 16).

⁴⁶ The component parts of Wilkins’s real character were cut by Joseph Moxon (1627–1691), Royal Hydrographer and future F.S.R., who is thought to have used Wilkins’s work to justify his first forays into type-founding (Mores [1778] 1961, 34; Davis and Carter 1978, 357, 364–5). On the printing of Wilkins’s Essay, see Lewis (2002) as well as Dolezal and Risvold (2018). Cf. Lodwick, whose unpublished shorthand manual had been halted at the press in 1650 due to his inability to procure woodcuts (Morian to Hartlib, 21 January 1650 [N.S.]). A century prior, John Hart had reeled back his plans for a reformed alphabet so that it could be more cheaply committed to press (his initial plans involved a custom font in imitation of secretary hand, but he eventually settled for an expanded set of italic letterforms) (1551, 225; 1569, sig. K4^v). For more on the trends and challenges of committing unusual letterforms to print, see Lucas (1999, 2000, 2018).

⁴⁷ ‘sicut delectum feci sonorum simplicissimorum & maxime euphonicorum; sic etiam Characteres erunt simplicissimi & figura [sic] pulchræ & uniformes: nullæ erunt caudæ dependentes, nulli apices eminentes’ (1661, 17).

⁴⁸ Indeed, Wilkins levelled criticism of his own on the look and function of other writing systems: ‘The Hebrew Character, as to the shape of it, though it appear solemn and grave, yet hath it not its Letters sufficiently distinguished from one another, and withall it appears somewhat harsh and rugged. The Arabic Character, though it shew beautiful, yet is it too elaborate, and takes up too much room, and cannot well be written small. The Greek and the Latin are both of them graceful and indifferent easie, though not without their several imperfections’ (1668, 14).

⁴⁹ This idealistic goal was curbed for practical concerns (the harder the language, the less likely it was to be taken up and disseminated), and in reality all language planners privileged the Anglophone and Latinate worlds above all others, not only in phonetics and orthography, but in lexicon, grammar, etc. Dalgarno, for one, was pleased with the idea that if his system did take off, certain Anglophone aspects would be felt by all ‘as a badge of its origine’ (Cram & Maat 2001, 383), and therefore hesitated little over his decisions to, for example, default to the English norm in cases such as weights and measures (1661, 89–90).