Praise, My Soul, the King of Heaven – or was it Hivven?

In eighteenth- and early nineteenth-century Britain, hymns and poems generally rhymed "heaven" with "given" or "forgiven", suggesting an alternative popular pronunciation of the word itself. "Heaven" also had two specific features which limited the scope for poetic pairings. Purists insisted that the second part of the word should be clipped or swallowed, making it almost monosyllabic.  In addition, it referred to a majestic afterlife, and was treated as a taboo word that required to be rhymed in a respectful, even reverent manner. These constraints retreated in Victorian times. The spread of basic literacy seems to have encouraged people to voice words as they saw them on the page, and "heaven" became bisyllabic. At the same time, less literal interpretations of the Bible encouraged a relaxed and hedonistic use of the concept, whimsically reflected in popular verse. In the absence of sound recordings before the invention of Edison's phonograph, it is impossible to know precisely how any words were pronounced. Hence this exploration of "heaven" is necessarily inconclusive, but it strays into some curious byways in its search.   

It is more than sixty years since I was last compelled to attend formal religious worship at a school assembly.  The exposure inoculated me against religion, while helpfully providing me with enough basic information to identify its manifestations in the past, a useful tool for a historian.[1] But I have been left with an enduring affection for soaring, roaring noise of some of the more expansive Anglican hymns, even those – like Onward Christian Soldiers! – that are burdened with embarrassingly boneheaded lyrics. The opening verse of one of them, Praise, My Soul, the King of Heaven, has long puzzled me: "Praise, my soul, the King of heaven / To his feet thy tribute bring. / Ransomed, healed, restored, forgiven, / Who like me his praise should sing." "Ransomed, healed, restored, forgiven" is splendid. Each of the four words requires distinct articulation, breaking the flow of the line in the reinforcement of their individual assertion. Together they combine to summarise key elements in the Christian message of salvation. But why is there this awkward near-miss of a rhyme between "Heaven" and "forgiven"?[2] In my autumn years, it has occurred to me to explore the mystery. This essay is the product of a series of more-or-less random raids into various source materials and can hardly claim to be an authoritative review. Nonetheless. I hope the journey may interest others who share my fascination with words.

It seems hard to believe that this anomaly was caused by the versifying incompetence of the Reverend Henry Francis Lyte, who penned the words of Praise, My Soul, the King of Heaven in 1834. Born in Scotland in 1793, he was educated in Ireland before making his career in the Church of England. As a student at Trinity College, Dublin, he had won poetry prizes. He may therefore be assumed to be someone who had experienced the regional variations in English pronunciation, and who well understood how rhyme reinforced the impact of metre. Above all, he was a fertile and successful hymn-writer, being remembered perhaps above all for the legendary anthem, Abide with Me, arranged in impactful and sometimes moving couplets, with the exception of a single eye rhyme that paired "word" with "Lord".  It is true that in later verses of Praise, My Soul, the King of Heaven, Lyte paired with "favour" with "ever" and, in a verse generally omitted, "flourish" with "perish". But, surely, in the thundering opening verse of his great hymn, Lyte would have avoided any jarring combination that would have deterred congregations from lifting their voices to the Almighty?

In deciding to explore the pronunciation of "heaven", I have attempted to deal with my lack of technical competence in phonetics by using primitive renderings in square brackets to indicate how words were spoken.[3] Broadly, there are two main potential explanations for the Praise, My Soul, the King of Heaven rhyming mystery. The first is that Lyte expected that congregations would sing "heaven" as [hivven], and that this usage was common in hymnology until mid-Victorian times. Evidence for this as the accepted pronunciation, probably more precisely in the clipped form [hivv'n], with a swallowed second syllable, can be traced in poetry back to the mid-seventeenth century. However, by about 1780, it is possible to identify [hevven] (or the abbreviated [hevv'n]) as an alternative spoken form, which may suggest that it was in fact the default version. I have traced it occasionally in poetry until about 1820 but, in the next half century, the pronunciation that we use today seems to have made little or no headway either in hymns or verse. Perhaps it had gone underground and was hiding in plain sound for, sometime around 1880, [hivven] went into rapid retreat, and [hevven] emerged not only triumphant but also fully bisyllabic.[4] If this second explanation ([hevven] or [hevv'n]) is valid, then other reasons must be sought for the persistent use of rhyming words such as "given" or "forgiven" that create the impression that "heaven" was [hivven] or [hivv'n].  These might include the need to pair with terms that also had a clipped second syllable, or the imperative in an age that venerated the vocabulary of holiness with employment of suitably respectful rhymes: hence "given" and "forgiven" conveyed a religious message, while "seven" and "leaven" conveyed no aura of sanctity. Although the [hivven]-style rhymes dominated the two great Anglican hymnals, published in 1861 and 1870, linguistic and cultural change quickly followed. The advance of literacy among the masses may have encouraged a popular process, however much frowned upon by purists, of voicing words as they were spelt, so that it became logical to pronounce "heaven" as two roughly equal syllables. More generally, the later Victorians tended to be more relaxed about their religious beliefs, or least less shocked by workaday allusions to sacred terminology. Bisyllabic and at least partly demythologised, "heaven" could now be comfortably rhymed with "seven" and even roguishly paired with "Devon". The material is tantalising, but it can be entertaining. Nonetheless, there remains an imponderable problem about the two main bodies of evidence: the versifiers wrote as if they were seeking rhymes for [hivven]; the elocutionists insisted that the word was [hevven].[5]

Recovering lost speech It can be peculiarly difficult to pin down the pronunciation of contested English words before the invention of sound recording in late-Victorian times.[6] In her 1999 monograph, English Pronunciation in the Eighteenth Century, Professor Joan Beal noted that the relative paucity of studies of English speech in that era. The eighteenth century saw the adoption of more-or-less standardised spelling: the landmark here was Dr Johnson's Dictionary, published in 1755, with an enlarged third edition in 1773, although it was not the only such development.[7] Unfortunately, these tacitly agreed forms that were not always phonetic. One intractable reason for this lay in disagreement over the pronunciation of even basic words. Johnson related the tale of two of Westminster's most powerful orators, Lord Chesterfield and Sir William Yonge, who clashed over the rendering of an adjective so basic as "great": Chesterfield embraced the modern pronunciation ([grate]); Yonge insisted on [greet] and denounced the version we use today as an Irish barbarism.[8] Johnson himself retained a Midlands accent, evident in his use of [heerd] for "heard" and [poonsh] for "punch" (the alcoholic beverage). The establishment of an approximate consensus in regard to spelling was followed by the appearance of a series of pronouncing dictionaries, of which the works by Thomas Sheridan (1780) and John Walker (1791) remain particularly useful, while a revised reissue of the latter by B.H. Smart in 1836 gives useful glimpses of the way the language had changed over half a century. Sheridan and Walker did not always agree: Sheridan decreed that "geographer" began [jog], while Walker opted for two syllables, [jee-og].[9] Sheridan made "bomb" rhyme with "Tom" and "from"; Walker insisted upon [bum], which he also applied to the first syllables of "bombard", "bombardment", "bombardier" and "bombast". One might have expected that these pronunciations would have been phased out during the Napoleonic Wars, but they were still accepted by Smart twenty years after Waterloo.[10] More puzzling was Sheridan's verdict that "front" should rhyme with "font"; Walker riposted that [frunt] was "almost universally" adopted by "custom" – although he was inconsistent in rejecting popular usage elsewhere. (The overlap between words spelt with –o– but pronounced with a short –u–, still a feature of modern English, is tangential to the question of [hivven] and is explored in an endnote, but it does illustrate the fluidity of vowel sounds in relatively recent times.[11] Precisely what the two compilers set out to achieve requires more detailed discussion below.

What is worth noting here is that Walker peppered his entries with opinionated comment, some of which provides insights into the spoken language of the time that – in all probability – would otherwise be entirely lost. Who would now know that "[e]a in fearful is long when it signifies timorous, and short when it signifies terrible, as if written ferful"? A particularly striking example is a question posed by Walker that, in the twenty-first century, would be regarded a straight choice between correct English and uneducated argot: should "my" rhyme with "fly" or "fee"? He insisted that both were correct in their own contexts, and explained their very precise differentiation: "this pronoun my, when it is contradistinguished from any other possessive pronoun, and consequently emphatical, is always pronounced with its full, open sound, rhyming with fly; but when there is no such emphasis, it falls exactly into the sound of me, the oblique [i.e. accusative] case of I. Thus if I were to say, My pen is as bad as my paper, I should necessarily pronounce my like me, as in this sentence pen and paper are the emphatical words; but if I were to say, My pen is worse than yours, here my is in opposition to yours, and must, as it is emphatical, be pronounced so as to rhyme with high, nigh, &c."[12]

The particular complication here was that pronunciation was changing in apparently random ways. Since Walker was determined that some changes should be blocked and others endorsed, he is not always a reliable authority for actual usage. There was, for example, a need for a common form of English speech that would provide a 'British' lingua franca to unite the English ruling class with Scots lairds and lawyers, while also providing a bridge to Ireland's (frequently absentee) Protestant elite. The subtitle of John Walker's Critical Pronouncing Dictionary and Expositor of the English Language announced that it featured "rules to be observed by the natives of Scotland, Ireland, and London, for avoiding their respective peculiarities", while the editor himself acknowledged that there was "scarcely any part of England remote from the capital where a different system of pronunciation does not prevail". Exaggerated modes of speech were sometimes adopted among the social elite to assert their superiority over the masses. Sometimes these spread down the social pyramid, only to be discarded at the top as outmoded. Thus forms that had been standard across all classes in one generation became despised as rustic peculiarities in the next.

From almost accidental asides, especially those of Walker, it is possible to detect that English pronunciation in the eighteenth century was sometimes very different from the standard forms of today. In many cases, it was also in flux, and the pronouncing dictionaries were actively engaged not simply in description but prescription, with compilers seeking to uphold their versions as the desirable norm.[13] A few examples may indicate this diversity. "About thirty years ago", Walker wrote in 1791, "the first syllable of Chamber was universally pronounced so as to rhyme with Palm, Psalm &c" – in other words, in conformity with the French word from which it was derived, "chambre". However, "since that time it has been gradually narrowing to the slender sound [of] the a in came, fame, &c. and seems now to be fully established in this sound" – a shift that he contested. By contrast, he deplored the early eighteenth century fashion in which "the word oblige was, by many polite speakers, pronounced as if written obleege, to give a hint of their knowledge of the French language". (Pope, for example, dismissed Joseph Addison as "by flatterers besieged / And so obliging that he ne'er obliged.") Then, in 1774, came one of the publishing sensations of Georgian Britain, the appearance of Lord Chesterfield's Letters to his Son, which set out rules for "the Fine Art of becoming a Man of the World and a Gentleman". Chesterfield insisted that a "vulgar man" betrayed himself by saying that "he is OBLEIGED, not OBLIGED to you", adding that a man of fashion "takes great care to speak very correctly and grammatically, and to pronounce properly; that is, according to the usage of the best companies [i.e. social circles]".[14] Chesterfield had written in 1749, twenty-five years before his advice saw the light of day. Walker took up the tale. In the interim, [obleege] had become "so general, that none but the lowest vulgar ever pronounced it in the English manner; but upon the publication of this nobleman's letters..., his authority has had so much influence with the polite world as to bid fair for restoring the i [I], in this word, to its original rights; and we not unfrequently hear it now pronounced with the broad English i in those circles where, a few years ago, it would have been an infallible mark of vulgarity." Yet, as late as 1854, Punch could claim that Lord John Russell – born the year after the publication of Walker's dictionary – still pronounced the word as used "obleege".[15]

Another anomaly that seemed to be in retreat was the –ar– sound in "clerk" and "serjeant". "Thirty years ago [i.e. around 1760] every one pronounced the first syllable of merchant like the monosyllable 'march', and as it was anciently written marchant. Service and servant are still heard among the lower order of speakers, as if written sarvice and sarvant: and even among the better sort, we sometimes hear, Sir, your sarvant; though this pronunciation of the word singly would be looked upon as a mark of the lowest vulgarity." Walker recognised that "Derby" and "Berkeley" "still retain the old sound, as if written Darby and Barkeley: but even these, in polite usage, are getting into the common sound, nearly as if written Durby and Burkeley. Since the shift tended "to simplify the language by lessening the number of exceptions", he felt that "it ought certainly to be indulged".[16] His prediction would be realised in North America but not in Britain, where, half a century later, the preposterous neologism "Varsity" would emerge to describe Oxford and Cambridge.[17] Walker commented on other fads, such as the mangling of "gold" to rhyme with "fooled" – a trend that ultimately petered out – but there were undoubtedly contested pronunciations on which he did not comment. For instance, there is evidence that "general" was widely rhymed with "mineral" – as it would be almost a century later in the famous patter song in The Pirates of Penzance (1879): "I am the very model of a modern Major-Gineral / I've information vegetable, animal, and mineral". W.S. Gilbert even adjusted the spelling, and the ingenuity of his pairings – lot o' news / hypotenuse, Zoffanies / Aristophanes – confirms his intention to use an exact rhyme.[18]

This attempt to eavesdrop on the lost voices of Georgian Britain should be enough to illustrate that there remain considerable areas of uncertainty about how people spoke, making it less surprising than it might seem that a query could be raised about a term so basic to the national culture as "heaven". It is also evident that the language was changing in ways that even contemporary experts could not comprehend. For instance, just seventeen years separated Chesterfield from Walker and yet, in an era with no cinema or broadcasting, an apparent vowel shift had taken place in "obliged" entirely through word of mouth communication. Changes could also seem random: [marchant] had fallen from favour, but [Darby] would endure.

The process continued. In his 1836 edition, Walker Remodelled, B.H. Smart commented on pronunciation changes since the original volume had appeared. Exotic vegetables had become linguistically naturalised: "No well-taught person, except of the old school, now says cow-cumber or sparrow-grass, although any other pronunciation of cucumber and asparagus would have been pedantic some thirty years ago." Eighteenth-century Londoners ("and those not always of the lowest order", Walker had disapprovingly noted) had been notorious for confusing –v– and –w– at the start of words. "The diffusion of literature among even the lowest classes of the metropolis, renders it almost unnecessary to speak now of such vulgarisms", he reported, with evident approval.[19] In fact, arguably the biggest event in popular literature in 1836 was the publication of a sensationally successful debut novel called The Pickwick Papers, whose Cockney characters peppered its pages with "wery" for "very", and even threw in an occasional allusion to "weal pie" and "wentilation". Perhaps Charles Dickens was mining a vein of nostalgia for a recent, but lost, past, although Punch in 1843 could still celebrate "the friendly interchange of the v's and w's which so pre-eminently distinguishes the Cockney dialect".[20]

Smart even attempted to identify the shakers and movers who drove change in acceptable forms of speech: "they who commence these changes, however useful and necessary in their calling, are for the most part the smaller literati of the country, – they who attend more to manner than to matter, and love to lead the fashion in words, as others love to lead it in dress." He did not hold such people in high regard: "[t]o dispute the old and to settle a new pronunciation of a word is a task exactly suited to some abilities, which are not capable of a much higher flight".[21] This was not altogether persuasive. He did not explain how these minor literati were capable of influencing popular speech among a people notorious for robustly philistine attitudes, nor was it obvious that such obscure personalities could wield more influence than the generation of great poets that was coming to a close as Smart wrote.  Samuel Rogers, himself a poet, remarked shortly before his death in 1855 that it was "curious how fashion changes pronunciation. In my youth everybody said 'Lonnon', not 'London'." By "everybody", Rogers – a former City banker – meant the metropolitan elite: he cited Charles James Fox as an example. By late-Victorian times, 'Lunnon' (the vowel sound became more rustic) was associated with yokels on sleepy villages.[22] In the eighteen-thirties, the young men of Balliol College, Oxford were amused by the "old-fashioned pronunciation of certain words" in sermons delivered by their Master, Richard Jenkyns, who used a long –a– sound in "rather" [rayther] and made "wounded" rhyme with "hounded" ("like 'wow' in 'bow-wow'"). Jenkyns, it should be noted, was only in his mid-fifties, an indication of how speech could mutate within a generation.[23] Half a century later, antique forms still lingered. "Even now", commented a newspaper report in 1887, "many old-fashioned people pronounce 'dome' as 'doom' and 'Rome' as 'Room'."[24] While my survey of potential sources is anything but exhaustive, I am struck by the almost casual way in which evidence survives for such changes. For instance, in the eighteenth century, it seems that the second syllable of "balcony" had a long –o– sound, faithful to its Italian derivation. Its transition to the shorter form of the vowel that we use today is commemorated because Samuel Rogers complained that the change in usage "makes me sick".[25] Many other similar shifts may well have been taken for granted.

In all of this uncertainty, there are two potential sources for establishing pronunciations: verse, in the form of poetry and hymns, and the pronouncing dictionaries already mentioned.

Rhymes and rules It might seem that poetry and hymns offer an absolutely reliable source for past pronunciation: English culture has generated a great deal of both and, unless both rhyming words have somehow become changed – which seems unlikely – then versification should unlock the ways in which people voiced their words in past times. A well-known example comes from Alexander Pope's delighted portrayal of the domestic life of Queen Anne, sovereign of England, Scotland and Ireland, as she relaxed at Hampton Court: "Here thou, great Anna! whom three realms obey, / Dost sometimes counsel take – and sometimes tea." It is taken as proof that the infused beverage from China that had taken Britain by storm during the previous fifty years was known in 1714 as [tay]. However, there is a complication. One incidental nuisance associated with the orthography that became settled by the eighteenth century is that it has enabled English poetry to makes use of an irritating device called the eye rhyme, which allows lines to end with words that look alike ('love' and 'move' are examples) but which jolt the reader with their disruptive dissonance. Poets who employed eye rhymes were neither incompetent nor out to create confusion – and it was not a new device. Commenting on Shakespeare's use of them in Venus and Adonis, Beal remarked that he "rhymed war with jar and warm with harm ... because he was writing within a tradition which demanded end-rhymes and because those words fitted in with the theme of his poem".[26] This is reassuring since, in Venus and Adonis, the Bard also paired "heaven" with "even". It seems unlikely that anybody who spoke English as a first language ever pronounced the word as [heeven], but the existence of the eye-rhyme convention can complicate attempts to interpret other pairings: where there appears a little-used word, such as "leaven" or "seven", are we looking at evidence for [hevven] or should the usage be dismissed as a vague and rogue approximation? (My dismissal of [heeven] is disgracefully anglocentric: the Scots National Dictionary has examples of a firmly bisyllabic "heevan", but these do not seem to explain or justify the eye rhymes.)

However, these concerns would seem to apply mainly to what might be called "high" poetry, the solemn and evocative verse that was read silently or aimed at very small groups of devotees. Hymns were intended to be sung by congregations on a lift-up-your-hearts basis, while popular verse – increasingly important by the second half of the nineteenth century, especially through Punch – depended for its effect upon the impact, and sometimes the ingenuity, of rhyme. (Here, as will be discussed, there was the limiting factor of the sensitivity in the handling of sacred concepts which – as with the limited range of options faced by Shakespeare in Venus and Adonis – required "heaven" to be paired with appropriately respectful terms.) Thus, in the absence of sound recordings before the late nineteenth century, it is impossible to be sure whether the far more frequent rhyming of "heaven" with "given" and "forgiven" than with "leaven" and "seven" was because the word was pronounced [hivven] or because the former were simply more tasteful than the latter.

Unfortunately, in relation to the pronunciation of "heaven", the two basic sources provide different answers. The hymns and most of the poetry until the mid-nineteenth century point to [hivven], but the pronouncing dictionaries come down on the side of [hevven]. There is some literary evidence in support of the use of [hevven], at least as a minority variant, in the early nineteenth century, and the rapidity with which popular verse endorsed this modern form after about 1880 suggests that it had remained in common use, even though mid-Victorian sources rarely seem to have recognised it. Given the lack of clarity, I proceed from this point first by examining the case for [hivven], initially through hymns and then by dipping into poetry, before considering the evidence for [hevven] provided by occasional literary exceptions. It is at this point that it will become useful to return to the question of the pronouncing dictionaries.

The case for [hivven]: the Victorian hymns It speaks volumes for the fractured theology of nineteenth-century Anglicanism that the Church of England was unable to agree upon a uniform hymn book. Hymn-singing had only taken root in most churches from about the eighteen-thirties: paradoxically, it was an import from the Dissenters popularised by the Oxford Movement. Hymns Ancient and Modern (1861) was intended to supersede the bewildering variety of hymnals that had sprung up. A collective project and "of moderate Tractarian  inspiration", its sometimes suspect doctrines produced E.H. Bickersteth's rival Hymnal Companion to the Book of Common Prayer nine years later: the title, with its Protestant overtones, conveyed an implicit 'beware of inferior imitations' warning.[27] 

My survey of Hymns Ancient and Modern mostly sought to identify specific examples of heavenly rhyme, but my incidental statistical efforts definitely reinforce Lyte's use of "heaven". I noted thirteen examples that rhymed with "given", six with "forgiven" and one with "driven": unless these words were voiced as [gevven] and so forth, which seems unlikely, then either "heaven" was pronounced as [hivven] (if only for ecclesiastical purposes), or hymn-writers were casual and cavalier in their choice of vocabulary. The usage can be traced back to the eighteenth-century Methodist origins of modern hymn-singing. Thus Charles Wesley's Forth in Thy Name, O Lord, I go (1749) proclaimed: "Fain would I still for Thee employ / Whate'er Thy bounteous grace hath given; / And run my course with even joy, / And closely walk with Thee to Heaven." The early nineteenth century produced more examples from other strands of Protestantism. In New Every Morning Is The Love (1822), John Keble sang of: "New perils past, new sins forgiven,/ New thoughts of God, new hopes of heaven." Five years later, Reginald Heber wrote that "God, Who madest earth and heaven, / Darkness and light; / Who the day for toil hast given, / For rest the night." It is worth noting here that the Anglican Church benefited from some talented female hymnodists. In 1841, Sarah Flower Adams wrote the words of Nearer, My God, to Thee, sometimes claimed to have been played by the ship's band as the Titanic sank in 1912. "There let the way appear, steps unto heaven; / All that thou sendest me, in mercy given" (some versions prefer "heav'n" and "giv'n"). That same year, Frances Elizabeth Cox translated a German hymn, O Let Him Whose Sorrow, which used a different rhyme but the same vowel sound: "Raise thine eyes to heaven / When thy spirits quail,/ When, by tempests driven, / Heart and courage fail." Shortly before the publication of Hymns Ancient and Modern, Catherine Winkworth produced an English translation of one of the most magnificent of all German hymns, Now Thank We All Our God: "All praise and thanks to God / The Father now be given, / The Son and Spirit blest, / Who reign in highest heaven / The one eternal God, / whom heaven and earth adore; / For thus it was, is now, / and shall be evermore." Given, heaven, adore, evermore: the rhymes seem to dictate the accompanying pronunciation.[28]

My survey of Bickersteth's Hymnal Companion to the Book of Common Prayer is more brutally arithmetical. I count 65 hymns which contain rhymes for "heaven". Sixty of these point to [hivven] as the pronunciation. "Given" scores 41 and "forgiven" 16, while "striven" is used twice and there is a single exotic appearance for "thunder-riven". Of the five outliers, only one, a "seven", might point to the version of "heaven" in modern speech. Three others – two examples of "even" and a puzzling "engraven" – may be dismissed either as eye rhymes or eccentricities. The final example paired "heaven" with "Stephen", apparently an allusion to the first Christian martyr. It appears in a hymn by Charles Wesley (Head of Thy Church Triumphant) and may reflect an eighteenth-century rendering of the saint's name. Stephen was certainly not a widely used forename in Georgian Britain, and it is possible that its pronunciation was influenced by the short vowel sound in the German, Stephan. It can hardly undermine the overall conclusion derived from both hymnbooks, that when mid-nineteenth-century Britain hoped for paradise, they sought the road to [hivven].[29]

Something should perhaps be said about the order in which "heaven" and its putative rhyming words appear in various hymns. In the examples quoted above, "heaven" sometimes appears first, sometimes second. It might be reasonable to assume that, in the case of Praise, My Soul, the King of Heaven, "forgiven" was simply used as a conveniently anticlimactic way of rounding off the connection. However, to borrow colloquial terminology, where short –i– words are used as the warm-up leading to "heaven" as the punch line, we might more reasonably assume that the culmination was definitely intended as an exact rhyme. This seems to be the case in some eighteenth-century hymns, for instance John Morison's celebration of Epiphany, The People That in Darkness Sat: "For unto us a child is born, / To us a Son is given, / And on his shoulder ever rests / All power in earth and heaven." Similarly, Charles Coffin's How Blest Were They Who Walked in Love hailed the commitment of true believers: "To purer joys their hearts were given; / The better land they sought was Heaven." Yet the same apparent attempt to focus impact upon a [hivven] pronunciation can be found later, as in Keble's offering of 1822 in New Every Morning is the Love, quoted in the previous paragraph. Well into Victorian times, Winkworth's 1858 translation from the German, Christ the Lord is Risen Again!, used the same order of words, although her matching of "abroad" with "restored" may cast some doubt upon her commitment to strict accuracy: "Now he bids us tell abroad / How the lost may be restored, / How the penitent forgiv'n, / How we too may enter heav'n".

[Hivven] in verse: a brief dip into English poetry It has to be recognised that one of the first major poems in more-or-less recognisable English points strongly to the pronunciation of "heaven" as [hevven], as might be expected from its derivation from the Old English "heofen". John Gower's Confessio Amantis was written about 1390 and – unlike, it would seem, many later writers – he was not shy about using the term as an end-of-line rhyming word. Gower frequently paired "heaven" with "even[e]", which may have been an eye rhyme, but he also relied upon words that suggest the modern pronunciation: "seven" appears at least ten times, and he also called upon "el[l]even" and "levene" ("leaven"). Notably, Gower had access to two convenient rhyming words that later ceased to be current: "steven[e]" and "sweven". There are at least seven examples of "steven", which meant "voice", and two of "sweven", which meant "dream" and might convey the idea of a trance or vision. Both retain a place in the Oxford English Dictionary but are designated as obsolete: both are indicated as having a short –e– in their first syllable.

At this point, my trail runs cold for a quarter of a millennium. This is not to say that evidence does not exist, although it is fair to point out that both Shakespeare and Milton wrote extensively in blank verse. Yet, by the second half of the seventeenth century, English poetry seems to validate a pronunciation as [hivven]. If this was indeed the case, what was happening? It is well established that the English language experienced a vowel shift in the fifteenth and sixteenth centuries, sufficient to create a gulf between us and the world of Chaucer and Gower that helps to make Shakespeare and his contemporaries the starting point of a comprehensible literary tradition. But this sound shift affected the long vowels and could not plausibly explain a swing from [hevven] to [hivven] – if, indeed that is what took place.[30]  Most linguistic studies accept that the growth of London influenced the shaping of English: a population of around 50,000 in 1400 would pass half a million in the late seventeenth century. The city is assumed to have drawn in migrants from further afield, creating a babel of dialects that fused into new forms of speech. Unfortunately, raw demographics alone do not help much in this enquiry. There may be some encouragement to be found in the evolution of the place-name "Stepney" from its original "Stybbanhythe", the landing place of an unidentified Stybba: "Stib-" and "Steb-" forms coexisted throughout the Middle Ages, with "Stepney" first appearing as the telescoping of the middle syllable in 1466 made possible the softening of the consonant.[31] Yet even if we were to allow this probably isolated place-name example to persuade us of an overlap between short –e–  and short –i–  in early metropolitan English, it would still need to be explained how these demotic examples managed to register in poetry and hymnody. 

The Cavalier poets of the mid-seventeenth century provide some useful examples. True, in his Hesperides, Robert Herrick employed three rhymes which each point us in different directions. "And when those clouds away are driven. / Then will appeare a cheerfull Heaven" points to [hivven], but he also used one "even", while hailing the beauty of a bride in what sound like modern tones: "Reaching at heaven, / To adde a nobler Planet to the seven". (The significance of "seven" is discussed below.) However, his contemporary Edmund Waller provides statistical reinforcement for [hivven], with eight examples of "given", one of "driven" and a sole intrusion of "seven". Richard Lovelace also used "given" as a rhyme on three occasions. Another Royalist writer, the famous angler Izaak Newton (died 1682), wrote of Richard Stibbes, a Cambridge divine of half a century earlier: "Of this blest man, let his just praise be given, / Heaven was in him, before he was in Heaven." From a slightly later generation, John Dryden defended the imposition of religious uniformity in his 1683 Religio Laici: "some Rules of Worship must be given, / Distributed alike to all by Heaven." (As Poet Laureate, it was his job to equate Heaven with the government.) Dryden also rhymed "heaven" with "forgiven", all of which tends to prove that [hivven] parallels were the well-established and basic usage, not some eighteenth-century affectation.

Alexander Pope seems to have remained faithful to [hivven] throughout his poetic career. In Thebais of Statius, which he translated in 1703 (and possibly earlier), he wrote of his subject: "Now by the fury of the tempest driv'n, / He seeks a shelter from th' inclement heav'n". Almost four decades later, in his Epilogue to the Satires (1738), he paid an often-quoted gracious tribute to an Irish bishop: "Manners with candour are to Benson given; / To Berkeley every virtue under heaven".[32] Pope helped Robert Dodsley establish himself as a bookseller, publisher and poet. One of Dodsley's lighter offerings was a roguish dialogue between the innocently outspoken Sylvia and a venerable clergyman – so venerable, indeed, that he was a Dean. "Cries Sylvia to a reverend dean, / What reason can be given, / Since marriage is a holy thing, / That there are none in heaven?" When the Dean chauvinistically suggests that marriage is impossible because there are no women in Paradise, Sylvia retorts that perhaps the problem is that an absence of priests makes it impossible to conduct weddings.

The Romantic poets also seemed generally united in their literary devotion to [hivven]. William Blake tended to eschew rhyming in his major works, but in 1802 slipped into the insensitive couplet: "My wife has no indulgence given / Except what comes to her from Heaven." That same year, Coleridge used the identical rhyme in his ode, Dejection. In 1808, Byron mourned the death of his dog, and could not see why so excellent a creature was denied eternal life: "While man, vain insect! hopes to be forgiven, / And claims himself a sole exclusive heaven." Wordsworth (in Ruth, 1815) praised the beauty of "a Youth to whom so much was given / So much of earth – so much of heaven". In Endymion (1817), Keats rhymed "heaven" six times, three of them with "given" and one with "striven", the participle also chosen by Shelley in his Ode to the West Wind (1819). (The other Keats examples from Endymion are discussed below.) The same pronunciation can be traced in later phases of Romanticism. Thus Emily Brontë in On a Wreath of Snow (1837): "O transient voyager of heaven! / O silent sign of winter skies! / What adverse wind thy sail has driven / To dungeons where a prisoner lies?" (Brontë herself was the prisoner – "My heart was weighed with sinking gloom" – who was cheered by the winter whiteness of the Yorkshire hills: "driven" was an obvious term to apply to snow.) Tennyson paired "heaven" with "given" deep inside In Memoriam (1850). John Henry Newman frequently used "given", and there are also individual examples in his verse of "driven", "forgiven" and "unforgiven". In The Dream of Gerontius (1865), there are four examples of "given" and one of "unforgiven". The theologian Brooke Foss Westcott even double-rhymed with short –i– words: "A rainbow arch in mercy given, / Amid Time's storm-clouds tempest-driven, / To span the earth and rest in heaven – / Oh, such is life."[33] My examples skim the surface of Victorian verse, but enough has been adduced to suggest that the primary choice for rhyme both in the two great mid-Victorian hymnals, [hivven], was also standard usage in contemporary literature.

The case for [hevven]: the minority exceptions? Yet there is one powerful piece of evidence that points to [hivven] as a literary and liturgical aberration. It is necessary to decode Walker's personalised use of diacritical marks, but there can be no doubt that his Critical Pronouncing Dictionary of 1791 firmly insisted on [hevv'n]. Walker was severe in his condemnation of the "vulgar and childish" pronunciation of "heaven" as two syllables (an issue separately discussed later) – but he uttered no criticism that implied that an alternative vowel sound was employed in church services.[34] Should this be regarded as conclusive? In evaluating the pronouncing dictionaries as historical evidence, we may begin by echoing Dr Johnson's question: "What entitles Sheridan to fix the pronunciation of English?"  Johnson's specific objection to Sheridan was that he was Irish, but it is perhaps for the best that "Sherry" did not live to respond to Walker's dogmatism.[35] As Beal more gently commented on Walker's assertions: "we must always be wary of his tendency to impose regularity where it may not have existed. He was, after all, in the business of laying down strict guidelines for a 'standard' pronunciation".[36] Walker certainly had little reason to draw attention to alternatives to his own promulgations and, as he acknowledged, there was "scarcely any part of England remote from the capital where a different system of pronunciation does not prevail". In fact, he did admit that some words spelt as if with a short –e– were actually rendered with a short –i– and, in one case, he bowed to popular usage. It is striking to learn that so basic a word as "yes" was once pronounced as [yiss]. "This word is worn into a somewhat slenderer sound than what is authorised by the orthography; but e and i are frequently interchangeable, and few changes can be better established than this."[37] However, he took a stand in defence "yet" as [yett]. "The e in this word is frequently changed by incorrect speakers into i; but though this change is agreeable to the best and most established usage in the word yes; in yet it is a mark of incorrectness and vulgarity." He also complained that musicians rendered "clef" as [cliff], a surprising usage which, unlike the early eighteenth-century versions of "chamber" and "oblige", can in no way be explained by its French derivation. Perhaps the vertical nature of the musical symbols somehow invoked the image of a cliff. Walker contemptuously attributed the distortion to craft tribalism.  "It is a common fault of professions, liberal as well as mechanical, to vitiate their technical terms." It was a mispronunciation, as he curtly pointed out, that lacked even the defence of brevity.[38] In addition to his explicit compromise on "yes", there were two other noteworthy examples where Walker silently endorsed the sound shift from short –e– to short –i–: "pretty" as [pritty] and the anomaly still taken so much for granted that few speakers of the language are perhaps even conscious of it – [inglish] for "English".[39]

Given that Walker did not contest every variant pronunciation, his lack of comment in endorsing "heaven" as [hevven] was perhaps understandable.[40] In 1791, hymn-singing was primarily associated with the Methodists: Walker had spent the early part of his career on the stage and was unlikely to have had much contact with Wesleyan revivalism. It was safe enough to scorn the hapless ignorance that mangled the first syllable of "engine": "Pronouncing this word as if written ingine, though very common, is very improper, and savours strongly of vulgarity."[41] People who talked about engines worked in the mining industry or served in village fire brigades. If [hivven] was indeed preferred from the pulpit, it was probably more tactful to pass over it in silence.[42] An enthusiast for the spoken language, Walker was not much interested in poetry. However, he did quote one revealing couplet, criticising Alexander Pope for truncating "to" to "t'" as part of the infinitive: "Say what the use, were finer optics giv'n, / T' inspect a mite, not comprehend the heav'n?" His lack of comment perhaps constitutes a silent endorsement of a popular pronunciation.[43]

Publishing forty-five years later, Smart quietly ignored some of his forerunner's opinions: "engine" was now unambiguously "en-jin", while "yes" was pronounced with a short –e–. Yet he broadly endorsed Walker's concerns. The core problem, as Smart saw it, was that while English had just seven basic vowels, "these seven sounds, modified, qualified, and compounded, give existence to fifty-five varieties of vowel sound". Like Walker, he recognised and accepted some degree of overlap between short –e– and short –i–.  Indeed, there were many cases where "there would be stiffness and pedantry in scrupulously adhering to regularity; for instance, whenever letter e makes an additional syllable with s in forming the plural, or the genitives of nouns, or the third person of verbs; as in box-es, fa-ces, Geor-ge's, he practis-es, he deba-ses" and "likewise in the last syllable of helmet, poet, linen, covet, &c., although marked in the dictionary for its regular sound." He too acknowledged that "custom" had changed the short –e– to a short –i– "in English, and in pretty", adding that this was also the case "in the musical term clef", which Walker had also briefly mentioned.[44]  

There was, too, one other possible consideration that constrained all the compilers of pronouncing dictionaries. Heaven was invoked in English churches on the Sabbath, but it was also an everyday term in popular exclamation. "Good heaven!" (usually singular in those days), "gracious heaven!", "thank heaven!", "heaven forbid!" – such expressions were used by the pure and demure heroine of Samuel Richardson's Pamela in the time of George II, and by Jane Austen's robust young ladies in Regency times two-thirds of a century later. Perhaps the pronunciation in this popular context was indeed [hevven].[45]

Yet if [hevven] was indeed the default pronunciation, it seems to have been rarely used in literature (and hardly at all in hymns). However, from the late eighteenth century, at least, there was a substratum of examples where poets paired "heaven" with "seven". Wordsworth in 1798 wrote of a little girl who told him about her two dead siblings. "'How many are you, then,' said I, / 'If they two are in heaven?' / Quick was the little Maid's reply, / 'O Master! We are seven'." In The Idiot Boy that same year, Wordsworth similarly used "eleven", although later he favoured "given" and "forgiven". William Mackworth Praed (who died, young, in 1839) used the "seven" rhyme in his love poem, A Chapter of Ifs: "If day were night, if six were seven, / Pain pleasure, monkeys men, / If thou wert worthy aught but Heaven – / I might forget thee then." Elizabeth Barrett Browning also rhymed "seven with "Heaven" in her impenetrable song to The Seraph and the Poet, which was not published until after her death in 1861. With typical exaggeration, William Blake (d. 1827) used "seven times seven" and "seventy-seven". Could it be that "seven" was also pronounced as [sivven] – possible, although perhaps less plausible with Wordsworth's use of "eleven"? Or was it that these "seven" rhymes represented near-miss versification, combinations that were designed to be approximate – hardly a compliment to some of Britain's greatest literary figures.

Here it is useful to consider the use of another word, "leaven", by Keats. Walker treated its first syllable as a short vowel, as we do today.  Its use in Endymion was contemptuously denounced by J.W. Croker, in a diatribe that bears both upon the pronunciation of "heaven" and the question of the appropriateness of the terms that were coupled with it. In an unimpressive segment, Keats wrote: "Be still the unimaginable lodge / For solitary thinkings; such as dodge / Conception to the very bourne of heaven, / Then leave the naked brain: be still the leaven..."  Croker, already a savage critic, exploded on encountering lodge, dodge, heaven and leaven in a work of poetry, insisting that "the rhymes when filled up shall have a meaning", a task beyond Keats who "has no meaning". Rather, he "seems to us to write a line at random; and then he follows not the thought excited by this line, but that suggested by the rhyme with which it concludes. There is hardly a complete couplet inclosing a complete idea in the whole book. He wanders from one subject to another, from the association, not of ideas but of sounds, and the work is composed of hemistichs; which, it is quite evident, have forced themselves upon the author by the mere force of the catchwords on which they turn."[46] Hemistichs are half-lines, and Keats had certainly crafted some awkward collisions in this extract.

Two conclusions follow from Croker's denunciation. First, he seems to have accepted that "leaven" was an acceptable rhyme for "heaven": indeed, his complaint was that Keats had allowed the similarity of sound to seduce him into a meaningless statement. Croker denounced Keats and Leigh Hunt as the "Cockney poets", and it is possible that his ire was stoked by a perceived lapse into London argot.[47] Second, the outburst was a reminder that traditionalists insisted that poetry represented attempts to capture truths in the magic of language, and was not simply about lumping together words of similar syllabic structure. Words associated with "heaven" had to convey respect, and those who employed them needed to be on their guard for problems of interpretation. Nineteenth-century Britain saw an in tense debate on the definition of the concept: was Heaven a place or an abstract state of existence?[48] Hence it is no surprise that the rhyming words were limited not just by sound but by relevance, with "given" and "forgiven", terms of salvation, were much more suitable than "driven" and "striven", which implied conflict.[49]

The complication seems to have been that the two most obvious [hevven] rhymes, "leaven" and "seven", were not particularly appropriate to celestial mystique, if Croker's criteria were applied. "Heaven" is a term of solemn purity, while, when not specifically referring to yeast, "leaven" conveys lightness, even dilution.[50] "Seven" was even more problematic. In Judaism and Islam, there are seven levels of Heaven, a structure that never made much impact upon Christian theology, although we still describe an ecstatic experience as "seventh heaven". In an era when astronomy overlapped with astrology, telescopes could spot seven prominent bodies in the sky – the Sun, the Moon, Mercury, Mars, Venus, Jupiter and Saturn – hence Herrick's celebration of a bride so gorgeous that she could "adde a nobler Planet to the seven". These seven celestial objects triggered the parallel idea that each had its own heaven, an essentially pagan notion that would certainly not have appealed to hymn-writers. In any case, the identification of Uranus in 1785 upset the arithmetic and further undermined the utility of the rhyme. It becomes possible to understand how "given" and "forgiven" could become so popular as pairings for "heaven", even if the word was pronounced as [hevven]: they conveyed the right message however much they might be second-best as versification. But can so many examples be explained simply as clumsy approximation? In fact, a formula could have been found to use "seven" as a protest against sabbatarian narrowness.  In a poem published in 1633, George Herbert challenged the  practice of confining worship to Sunday: "Seven whole days, not one in seven, / I will praise Thee. / In my heart, though not in heaven, / I can raise Thee."  The sentiment might have been inverted to make it more palatable to British Protestants who firmly limited their devotions to one day a week, versified to claim that the joy of praising Heaven made Sunday the best day of the seven. The fact that no popular eighteenth- or early nineteenth-century hymn appears to have embodied such an idea again suggests that the priority of hymnists was to find rhymes that paired with [hivven]. In the event, Herbert's poem made only a late and marginal entrance into hymnbooks, as King of Glory, King of Peace.

It is also worth noting that "heaven" seems to have been used relatively sparingly as a rhyming word. In some respects, this is surprising, since other key Christian terms faced even greater challenges in the search for appropriate pairings. There were almost no convenient matches for "Christ", "Jesus" and "saviour", while words that rhymed with key concepts such as "God", "church", "soul" and "sin" were either trivial or unfriendly – "cod", "hod", "nod", "pod", "sod" and "shod" contributed nothing at all to the adoration of the Almighty, a dearth that perhaps explains occasional desperate attempts to drag in "road".[51] Yet "heaven" was hardly overworked in their place. Even where the word completes a line, it does not always have a partner: an example is the gloriously repeated cry of "Bread of heaven, bread of heaven" in Guide Me, O Thou Great Redeemer, translated from its Welsh original in 1781. Remarkably, this resounding line has no rhyming counterpart in the hymn at all. Wordsworth's 1804 invocation of the impact of the French Revolution fifteen years earlier remains one of the most striking quotations from the Romantic movement: "Bliss was it in that dawn to be alive, / But to be young was very heaven!" It is tempting to assume that the message – and the poet's subsequent disillusionment – must have been rammed home in resounding rhyme. In fact, it is not so much a quotation as an excavation from a longer sequence, the line ending awkwardly with "oh, times..." – and the whole poem was crafted in blank verse anyway.[52] Perhaps the aura of pearly gates, clouds of glory and thrones of judgment was a deterrent, but there is surely also some sense that the more devout and respectful poets were reluctant to treat "heaven" as a mere gadget in a versifier's toolkit. This, of course, was an obstacle to the use of the term in popular verse, but publications like Punch overcame their reluctance as the nineteenth century wore on, presumably because their readers became less straitlaced about the pirating of religious vocabulary.

Later Victorian times The two great collections of hymns published in 1861 and 1870 confirmed the predominant practice of Britain's poets of rhyming "heaven" as if it rhymed with "given". Whether clergy invoked the concept and congregations actually roared it out as [hivven] it seems impossible to say: perhaps, like their modern successors, they were simply stuck with a fossilised eighteenth-century pronunciation.  By contrast, it can be reasonably argued that popular sources point to the affirmation of [hevven] as the mainstream pronunciation by about 1880, and it is to these that I now turn, concentrating especially upon sample surveys of Punch at intervals of about a decade.[53] No great claim is advanced for the literary merits of Britain's humorous weekly magazine; it is rather that the impact of its light verse, especially when it degenerated to doggerel, depended on the accuracy of its versification, making it a useful guide to contemporary pronunciation.[54]

Heaven was not a concept that had sat comfortably in early Victorian light rhyme. The literary gadfly Thomas Hood, who died in 1845, generated two contradictory examples. A [hivven] pronunciation is suggested by the use of "riven" in a sonnet the attempted to capture the unconscious thought-processes of a sleep-walker: "Methought – for Fancy is the strangest gadder / When sleep all homely mundane ties hath riven – /  Methought that I ascended Jacob's ladder, / With heartfelt hope of getting up to Heaven". But in his Ode to Rae Wilson (a Scottish critic whom he handled somewhat roughly), Hood adopted "leaven" for the same purpose: "My heart ferments not with the bigot's leaven, / All creeds I view with toleration thorough, / And have a horror of regarding heaven / As anybody's rotten borough." Another contemporary effusion is of interest in calling upon "driven", a matching word that conveyed little in the way of spirituality and hence was rarely invoked by hymn writers. In 1842, the radical MP Thomas Wakley opposed attempts to tighten the law of copyright in the interests of authors. His parliamentary speech could be paraphrased as expressing the view that any fool could write poetry, a view that was not popular in the scribbling community. No less a newspaper than The Times carried a lengthy piece of disrespectful doggerel, which included the couplet referring to "Thoughts that, like seeds at random driven / Across the sweet fresh breath of heaven".[55]

Although founded in 1840, Punch in its early years was markedly cautious in its versification, no doubt for fear of causing offence by seeming to mock a revered spiritual concept. Two decades later, on the eve of the publication of Hymns Ancient and Modern, Heaven was still not a subject for frivolity. Darwin's Origin of Species in 1859 and the blunt challenge to orthodoxy in Essays and Reviews the following year probably help to explain why the only two poetic examples it carried at that time use the term in the context of a virtual pantheon of British national heroes. The search for the expedition of Sir John Franklin, the Arctic explorer missing since 1847, had produced reports from Inuit that his men had perished in appalling conditions trekking across the frozen wastes. Punch, in 1859, drew some comfort from knowing that Franklin himself was already dead before the survivors had abandoned the ice-bound ships: "Not under snow-clouds white, / By cutting frost-wind driven, / Did his true spirit fight / Its shuddering way to Heaven." The attributed destination was safe enough: no charitable reader would question the hero's title to eternal bliss.[56] A few months later, the historian and politician Thomas Babington Macaulay was buried in Westminster Abbey, and Punch celebrated his interment among the statesmen lauded in his influential History of England: "Fitly his resting-place is given / With those great dead he loved so well. / Stand on his grave, and you may tell / The chief stars of our English heaven...".[57] Obituary verse in Punch was often artifical and usually declamatory: we may note that the choice of rhyming words in both tributes points to a [hivven] equivalent, thereby conforming to contemporary usage in poetry and hymns, but this would not necessarily indicate the use of such a stilted pronunciation in actual speech. However, it may be noted that, once again, both poems use "heaven" as the punch-line second part of the rhyme, which confirms the impression of a culminating push to [hivven].

By contrast, a decade later, there are definite signs that poetic conventions were in flux, and increased evidence that [hevven] was the standard pronunciation. Between 1869 and 1872, Punch rhymed "heaven" on fourteen occasions – three of them, remarkably, in a poem about the industrial Black Country around Birmingham: "given" (or "giv'n") was used six times, "driven" and "riven" once each. By contrast, "seven" appeared twice and there were two examples of a previously ignored word, "levin", a medieval term for lightning, which was invoked to describe both the Franco-Prussian War and the arrival in liberated Rome of Italy's king Victor Emmanuel. (This apparently unique use of "levin" suggests the idiosyncrasy of a temporary contributor.) An attempt was also made to weave "leaven" into verse denigrating the new concept of Home Rule, which is of interest because it was intended to be voiced in an Irish accent: "From Derry to Cork, and from Shannon to Liffey, / Give us Home-Rule, and wrong will be right in a jiffy.... / And d'ye think we'll be wantin' the blessin' of Heaven / Our Laws and our Press wid religion to leaven?"[58]

Thus, in this transtitional phase around 1870, there were eight rhymes with [hivven] and five with [hevven]. The remaining example, "even", formed part of an unusually clunking ditty, even by the undemanding standards of Punch.[59] Some contributors evidently favoured the traditional [hivven] rhyme – it is notable that an obituary tribute to the philanthropist Angela Burdett Coutts used "given" and "driven" to wish her on her way[60] – but others adopted a mixed approach. Punch itself complained about speech experimentation on the London stage. "Why do actors ordinarily say 'Skee]-y ' for sky, 'key-ind' for kind, 'Leeeew-cy' for Lucy, 'Ha-aven' for Heaven? ... This is not new; but no explanation has ever been given. A Pronouncing Dictionary for the Stage would be useful."[61] Yet it is striking that Punch invoked the older rhyme in saluting the momentous news that Queen Victoria had been proclaimed Empress of India. Addressing the sub-continent, it offered the assurance that "She holds your swarming millions now, but as a trust of Heaven, / To civilise and educate to her best teaching given". The awkwardness of the pairing is noteworthy.

Nonetheless, by the end of the decade, the transition was in full swing. In HMS Pinafore (1878), Captain Corcoran bemoaned a puzzling world: "Fair moon, to thee I sing, / Bright regent of the heavens, / Say, why is everything / Either at sixes or at sevens?" Perhaps Gilbert borrowed here from Praed, but there can be little doubt that he had the modern pronunciation in mind. Punch still occasionally paired "heaven" with "given",[62] but the traditional version was in retreat. By contrast, "heaven" was now sometimes rhymed with "Devon"[63] – not always as synonyms, but evidence both of the [hevven] pronunciation and the increasing trend towards two fully articulated syllables. In 1888, Punch deplored the decision to unveil a Trafalgar Square memorial to General Gordon without any dedicatory speeches: the martyred hero of Khartoum merited better: "For England, and Humanity, and Heaven; / The record of whose life should be a leaven / Of quickening greatness in a factious age / Of petty jealousies and Party rage."[64] A more serious offering followed two years later, as parts of Ireland faced the stark renewal of mass hunger, recalling memories of the terrible winter of 1847: "Who does not fear to speak of Forty-Seven, / When that same Shadow darkened all the isle? / Is it abroad once more? Avert it, Heaven!"[65] In 1895, it employed the same rhyme in a more cheerful setting. Well into middle age and still playing for Gloucestershire, the legendary cricketer, W.G. Grace, scored his hundredth first-class century. Punch rejoiced at his triumph: "nigh forty-seven – / There isn't a cheerier sight under heaven".[66]

There was, too, an even more remarkable piece of evidence indicating that, by the eighteen-eighties, [hevven] was not merely an alternative, but had effectively ousted [hivven] altogether in common and accepted speech (if, indeed, it had ever occupied such a position). Isaac Pitman had spent half a century arguing for spelling reform, inventing his own system of shorthand notation to support his campaign. Of course, there were arguments against any attempt to systematise English orthography. One was that regional and class differences in accents made it challenging to agree upon the precise phonetic rendering of any word. Another, more cynically practical, was that English spelling was so chaotic that reform would be a heartbreaking task. In 1887, Pitman countered this objection by proposing that the process be undertaken in stages, starting with the simplest changes – and he selected two examples that he evidently regarded as incontrovertible.  Step One in English spelling reform would involve "dropping the letter –a– in the combination –ea– when it represented a short vowel, as h-e-d, head; h-e-v-e-n, heaven".[67] This would seem to prove, incontrovertibly, that [hivven] had disappeared from the spoken language.[68]

Seven years after Pitman's proposal, a correspondence in The Times also incidentally referred to the pronunciation of "heaven". In the world of journalism, the late summer was known as the Silly Season: with the political world on holiday, newspapers scrambled to fill their columns and editors sometimes resorted to trivial issues. Thus when an Anglican activist who called himself "Lover of the English Language" proposed "a simple test" to winnow "our overgrown hymnals", The Times was happy to provide space, in the sure and certain hope that the idea would provoke an explosion of dissent. The plan was to discard hymns that used "false rhymes", such as pairing "God" with "road", "join" with "thine" or "bliss" with "less", the last of these offences an echo of Walker's acknowledgement in 1791 that short –e– and short –i– words were often conflated. While the writer felt "justified in looking with suspicion upon verses which show such poverty of idea or sanction such mispronunciations of words in our mother tongue", there was also an acknowledgement that "what seem to be vicious rhymes ... in some of our finest hymns" were in fact "words ... the pronunciation of which has changed in the course of time".[69] There was an element of contradiction here or, at the very least, some failure in understanding. It is clear, for instance, that Alexander Pope pronounced "join" as [jine] as in his 1709 Pastorals: "For you the swains their fairest flowers design, / And in one garland all their beauties join" – not to mention a conclusive range of examples in which he matched the word against "combine", "divine", "line", "mine", "shine" and "thine".

The most practical response to "Lover of the English Language" came from a clergyman who recalled, with exasperation, his days as a young curate, when he had been responsible for organising the musical element of the Sunday services. The incumbent's "peculiar theology condemned most hymns in the book", while the limited skills of the choir and the organist ruled out many others: "the churchwardens and some leading members of the congregation attacked the rest". "I do not hesitate to say that if we had then had any lovers of English about I should have had them waylaid after service. It would have been the last straw."[70] The ex-curate did not specify which hymns survived this minefield of prejudice and incompetence, but it is not difficult to understand why Praise, My Soul, the King of Heaven was so popular: a simple celebration of the love of God, it raised no controversial doctrinal issues, while even the clumsiest choristers could thunder out the Goss anthem.

Not surprisingly, the proposal to dump all hymns containing false rhymes triggered two detailed rebuttals, the first pointing out that the proposed rule of thumb would bar much-loved pieces of church music by some of its most spiritual hymnists. Hark, the Herald Angels Sing would go because Charles Wesley rhymed "proclaim" with "Bethlehem", a pairing that even Georgian congregations may have found a stretch. Isaac Watts had coupled "God" with "stood", and "come" with "home", thereby condemning O God, Our Help in Ages Past to the garbage bin. Widespread affection could not save New Every Morning is the Love, for John Keble had committed the named offence of matching "God" with "road", "love" with "prove" and – germane to this enquiry – "heaven" with "forgiven".[71]

Much the same ground was traversed by Lord Grimthorpe, the only contributor who signed his own name. An astronomically wealthy lawyer, prominent lay Anglican and ferocious controversialist, Grimthorpe ignored individual hymns to compile a wider lexicon of clashing rhymes, sufficiently broadly conceived to argue that the proposed principle of exclusion would cause too much carnage to work. One curious aspect of his list of condemned couplings is that many of them seem perfectly acceptable today. It seems difficult to understand why Grimthorpe objected to the pairing of "send" and "descend", or why he protested at rhyming "guide" with "ride" and "sanctified", "root" with "fruit", "eyes" with "surprise", "lays" with "praise" and "contained" with "arraigned".  I write in 2024, barely two lifetimes from the year 1894, but it would seem that the 55 vowel sounds that Smart identified sixty years earlier have continued to elide into smaller clusters. Even more surprising was the inclusion in Grimthorpe's list of howlers of "rate" and "great", surely an echo of the linguistic conflict that Dr Johnson had encountered while compiling his dictionary. The modern reader will feel more sympathy with his objection to the pairing of "God and abroad (which seems to be a very favourite one too)", and his bemusement at the coupling of "Pentecost" with "Ghost", presumably of the Holy variety. Crucially, too, he complained about the use of "riven" and "given" as rhymes for "heaven", which he condemned as "quite as bad as bliss and less".[72]

What is persuasive here is not, as Grimthorpe himself seemed to assume, the vituperative scorn of his invective, but the simple fact of his age. Born in 1816, he had survived Eton and graduated from Cambridge as far back as 1838. As a defender of traditional beliefs against Victorian novelties and a Low Church scourge of ritualism, he was thoroughly versed in Anglican liturgy and literature – and it was a familiarity that dated back over seventy years. Yet in his pointed dismissal of the predominant short –i– rhymes for "heaven" in the hymnbooks, he gave no hint that they had ever been valid in his own lifetime. It would seem that, although [hivven] still appealed to John Keble in the eighteen-twenties, the pronunciation had not been current during Lord Grimthorpe's childhood.

If indeed there had been a dramatic change in the pronunciation of "heaven" around 1880 (inconveniently soon after the publication of the two great hymnals) or, indeed, as early as the eighteen-twenties – it is hard to account for it, except as part of wider changes in English speech. Two superficially attractive explanations prove to be less persuasive when closely examined. One attractive possibility would point to the growth of literacy, with the concurrent development that people tended to say words as they read them: in 1965, Sir Ernest Gowers referred disapprovingly to "the speak-as-you-spell movement", and it seems that it was not new.[73] It will be recalled that, as early as 1836, Smart had attributed the elimination of the Cockney muddling of –v– and –w– to the "diffusion of literature among even the lowest classes of the metropolis". Here it is usual to make some airy reference to the Education Act of 1870, which established – or filled in the gaps – of a public system of primary schooling. But, when Pitman spoke in 1887, the oldest products of the new Board schools would have been barely thirty years of age, probably too young and certainly too socially unimportant to be responsible for a major shift in pronunciation. Nonetheless, there is evidence of gradually improving standards in literacy from the early Victorian period.  Marriage registrations showed that, in 1839, 59 percent of newlyweds in England and Wales could sign their names, a level of competence that was assumed also implied an ability to read. The percentage of signatures in marriage registers probably underestimated the numbers who were basically schooled, for it was possible to read without being able to write: in that same year, an informed observer estimated that barely one-quarter of the adult weavers in Bradford could write – but that two-thirds of them could read. There was also a gender disparity: half the men signed but only one-third of the women, which pointed to a larger percentage of households where newspapers and pamphlets might penetrate. Smart was right, too, about the impact of popular literature in the capital: 82 percent of Londoners who married that year could sign their names, three-quarters of the brides and over ninety percent of their husbands. It was also true that the trend was positive: by 1858, 68 percent of newlyweds could sign their names, and the gender gap was reported to be narrowing.[74] Nonetheless, we are hardly dealing with an educational tidal wave. Most people married in their twenties: an annual growth rate in the ability to sign between 1839 and 1858 of less than half of one percent was hardly revolutionary, and it would be further diluted within the wider adult community. True, the explosion of the press in the mid-Victorian period points to a growth in popular literacy, but the masses probably did not read publications that had much to say about Heaven.  

The second possible explanation would note the declining influence of the Church of England: if "hivven" was used by preachers, or at least thundered out by choirs, then presumably there were fewer people in attendance and likely to be influenced by the peculiarities of its vocabulary. (However, Grimthorpe's evidence, quoted above, would seem to suggest that any allusions to [hivven] had been phased out much earlier in the century.) Here, again, close examination of the available – and much disputed – statistics throws some doubt on the hypothesis. The religious census taken in 1851 – the only one of its kind – reported that, out of the total population of England and Wales of 17.9 million, just under 5.3 million took part in Anglican worship, while 4.5 million attended other Protestant churches. When allowance was made for the young, the sick and those who worked at weekends, the results suggested that about half the eligible population attended some form of religious service on the day the survey was taken. But what did the raw figures really mean? It could be argued that the number of churchgoers was really much smaller, since the truly devout attended both morning and evening services, and so were double-counted. On the other hand, while apologists accepted that at least five million people had skipped church or chapel services on census day altogether, they pleaded that those who failed to put in an appearance were not necessarily the same people every week: the army of absentees was "not always composed of the same persons .... The number of habitual non-attendants cannot be precisely stated from these tables". If that were so – and it did sound like special pleading – far fewer than five million English (and Welsh) people were entirely exempt from the vocabulary of organised piety.[75] The other problem with this argument is that here is not much evidence for any further decline in attendance at religious worship in the decades after 1851: the Nonconformists maintained their numbers until the end of the century, while a more energetic approach from Anglican clergy – coupled with a lavish building programme – may even have increased the overall total of their congregations (if not their percentage of the fast-growing population). Furthermore, it was generally agreed that the non-attenders of 1851 were overwhelmingly from the lower orders, and the working classes did not shape the formal language. However, it was evident that all the Protestant sects were weakest in urban areas, especially London and, in the long run, it was the cities that would determine standard speech.

Other possible explanations for the eclipse of [hivven] might include the dilution of 'Society', as new people with new money pushed their way into the social elite, a process that Anthony Trollope deplored in his 1875 novel, The Way We Live Now. Perhaps their impact disrupted the closed tribal codes: Gladstone (Eton and Oxford) mocked John Bright, the Quaker manufacturer from Rochdale, for using "transpire" as a synonym for "occur" / "happen".[76] And then there is the elusive notion of generational change. In precise reality, there is no such thing, for there can never be a frozen moment in time when one age cohort seizes the baton from its forerunner. Nonetheless, it can be argued that the later Victorians rejected many of the assumptions of their forebears, and perhaps may have discarded some of the forms in which those ideas were expressed. A possible example comes from the aristocratic Lyttelton family, although it must be admitted that they were ebulliently eccentric. In the mid-nineteenth century, their womenfolk pronounced "nasty" with a short –a–, rhyming it with "pasty" (as in "Cornish-", not "-faced": the Indian Prime Minister Shashtri is a reliable guide here).  The adjective was applied to anything to do with sex, and signalled that the subject was unmentionable. By the twentieth century, attitudes to sexual matters were becoming slightly more open, and both the word and – presumably – its specific pronunciation became associated with narrow-mindedness.[77] It may be that the later Victorians became embarrassed by the pearly gates and paradise version of the afterlife so long peddled by the churches: even Sydney Smith's friend who hoped to spend eternity "eating paté de foie gras to the sound of trumpets" became something of a figure of fun. Perhaps this explains why the older pronunciation was discarded in favour of the more prosaic and modern version. Of course, there is one other potential explanation for the apparently sudden triumph of [hevven] in the last decades of the nineteenth century. This would be that it had always been the standard pronunciation, and that all the poems and hymns that paired the word with "given" and "forgiven" were simply bending the vowel sound slightly to fit a sacred term with appropriately religious rhymes. But there are surely too many examples in contexts – notably the thunderous outbursts of hymns – to support such an interpretation, nor would it account for the occasional variants of "driven", "riven" and "striven". Moreover, we have the testimony of the stern Walker that "e and i are frequently interchangeable", even in instances where he strongly disapproved.

By the early twentieth century, "heaven" was no longer treated as a taboo term that merited appropriately pious end-of-line partners. It was thrown casually into romantic badinage: "... across the road, at No. 7, / Sweet Dolly grew neglected all her days / She'd sunny hair, and eyes as blue as heaven, / A dimpled smile, and pretty baby ways."[78] It was rhymed with "leaven" to capture a game of golf and, ingeniously, with "Oporto's juice of '87" to describe the quality of port served at dinners in the Inns of Court.[79] And there was no longer ambiguity about the way in which the word was pronounced, as Punch made clear when it saluted a New Year that, in reality, would turbulent and divisive: "Nineteen hundred and eleven! / Year with hope and promise gay, / Multiple of three and seven, / Rhyming perfectly with heaven".[80]

In the higher literary realms, there are also indications of change, although less certainty about the perfect rhyme for heaven. One complication here is that, while popular verse was now casual in its treatment of a formerly taboo word, some poets were deeply religious and made only rare and respectful reference to the concept of an afterlife. In an early verse, My Prayers Must Meet a Brazen Heaven, Gerard Manley Hopkins paired the term with "unforgiven" but, in 1883, two decades later, in The Blessed Virgin Compared to the Air We Breathe, he called upon the more modern pronunciation: "Blue be it: this blue heaven / The seven or seven times seven / Hued sunbeam will transmit / Perfect, not alter it." Five years earlier, in The Last Oracle, the notably less devout Swinburne had similarly invoked "leaven",[81] but Hopkins in particular was so reserved in his references that it is hardly persuasive to cite these few examples as anything more than a possible general indication of changing pronunciation. There were also overlaps in usage, perhaps to be expected in a medium that was not necessarily intended to be read aloud. Housman employed an eye rhyme in his evocation of Shropshire. Edith Sitwell, another devout believer, used both "riven" and "seven".[82]  In his 1899 Yattendon Hymnal, an attempt to sweep away Victorian cobwebs, Robert Bridges included George Herbert's poem of 1633 that had refused to confine worship to one day a week, thereby incorporating "seven" into a religious message. King of Glory, King of Peace was a very late addition to the repertoire. 

Two very different twentieth-century examples of "given" as a rhyme for "heaven" may perhaps be regarded as consciously intended to invoke an earlier time. Rudyard Kipling's cry of passion, A Nativity, was written in 1916, while he was deep in grief for the loss of his son on the Western Front. The opening lines of each verse, italicised in most published versions, briefly recite the story of the birth, death and resurrection of Christ. Each stage of the narrative is interrupted by the anguished cries of a mother who protests that her own son's death is without meaning. "A Star stood forth in Heaven; / The Watchers ran to see / The Sign of the Promise given – / 'But there comes no sign to me. (To me! To me!)'." No stone can be rolled away to reveal an empty tomb for this alt-Mary figure (in reality, Kipling himself): "For I know not where he is laid."[83] It is possible and indeed plausible to assume that Kipling used a rhyme redolent of Victorian hymnbooks to convey the idea of a traditional and superficial form of religion that could not provide the comprehension or the consolation required by the carnage of the First World War. Thirty years later John Betjeman similarly called upon what was now an antique pronunciation to offer a typically jaunty prediction of the grim, grey triumph of utilitarian and secular values in post-war nineteen-forties Britain: "...ev'ry old cathedral that you enter / By then will be an Area Culture Centre. / Instead of nonsense about Death and Heaven / Lectures on civic duty will be given; / Eurhythmic classes dancing round the spire, / And economics courses in the choir."[84] As a devout High Church Anglican, Betjeman did not lightly touch upon religious symbolism. I count only four other rhymes with "heaven", one of them another "given" (in 1966), but the remaining three, all dating from after 1954, conforming to the modern pronunciation – "seven", "1911" and – intriguingly – "Severn". It is difficult to believe that he could not have sketched his skittishly Orwellian vision by calling upon some rhyme – "leaven", with its connotation of emptiness, would have fitted appropriately.  Hence the appeal in the cathedrals poem to "given", in a purely functional sense referring to the delivery of lectures, was – as with Kipling in A Nativity – a nostalgic nod to the suffocating reassurance of Victorian piety. The examples cited are too rare and random to support any reliable statistical analysis, but they do confirm a trend towards [hevven] as the default pronunciation.

Aspiration and syllables  In 1878, a correspondent of The Times quoted verse in which the letter H complained at being dropped from the start of words: "... I have by you been driven, / From house, from home, from hope, from heaven".[85] As the "given" rhyme indicated, the complaint hailed from a previous generation, but it highlights one of the many problems of interpreting pronunciation from English spelling. Anyone attempting to decode the spoken from the written form – from Mars or, indeed,from Asia – would have few if any clues that the first letter is silent in "heir", "hour", "honest" and "honour", while it seems that it was not until well into the twentieth century that "hotel" became generally aspirated.[86] There was plenty of evidence of the dropped H in Cockney speech, and in the demotic dialects of south-eastern England generally, of which "'evin" (for "heaven") was an example.[87] Walker had insisted that there were "very few words in the language where the initial h is sunk": his list included "herb" and "herbage", "hospital", "humble" and "humbly", while he accepted that "humour" and "humorously" were rendered "as if written yewmour, yewmorous". However, he did not include "heaven" in his list of words where the omission of the initial aspirate was permitted.[88]

More enduring than the mystery of the silent H was the question of whether the word was fully bisyllabic.[89] From the eighteenth century at least, it is clear that "heaven" – whether [hevv'n] or hivv'n] – was neither one syllable nor two. Rather, it resembled a house with a verandah at the rear, the letter N being sounded as a kind of vowelless sub-syllable, sometimes described as the voice glide. "When Britain first, at Heaven's command..." – so begins Rule Britannia!, one of the United Kingdom's greatest secular anthems, and it is clear that James Thomson's verse of 1740 can only be fitted to Thomas Arne's stirring tune by truncating that key word.[90] Yet by the time Walker compiled his pronouncing dictionary in 1791, there was a tendency to add an emphasis that made the word into two equal parts. The root problem was that the English language contained other terms – such as "chicken" and "novel" – in which the second syllable was fully sounded, encouraging a tacit push towards standardisation with words that were similar in spelling. "This diversity in the pronunciation of these terminations ought the more carefully to be attended to," Walker insisted, "as nothing is so vulgar and childish as to hear swivel and heaven pronounced with the e distinctly [sic], or novel and chicken with the e suppressed."[91] Smart in 1836 regarded the phenomenon as an example of the dangers of a little learning, especially among Londoners. The Cockney "sinks, habitually and unwarily, the terminational vowel in chicken and Latin, novel and parcel. Correcting these, he carries his correctness so far as to make the vowel distinct in swivel and heaven, evil and devil."[92] An 1872 edition of Noah Webster's dictionary of American English complained that "of the most prevalent errors of the present day, especially among our clergy (for the laity have fallen into it much less), is that of pronouncing the words even (ev'n) evun, heaven (heav'n) heavun or heaven. ...  If the writer is correctly informed, it is never heard among good speakers in England."[93]

The syllabic issue provoked an interesting comment when it was raised in New Zealand in 1902. With the worthiest of intentions, a cultural group in the city of Dunedin organised a pronunciation competition, but the judges' decisions on range of disputed words quickly became a matter for public debate. Was it permissible to sound the final letter in "rendezvous"? "Should 'heaven" be "heav'n' or 'heav-en'?" This question provoked an ex cathedra response from local worthy Alfred H. Burton, who taught elocution and felt entitled (or perhaps compelled) to lay down the law: "'heaven' is pronounced 'hev'n' – almost a monosyllable.... Other instances would be 'seven', 'eleven', 'given', and 'striven', which word[s], I imagine, no one would pronounce in any other way than 'sev'n', 'elev'n', 'giv'n', and 'striv'n'." Burton may be regarded as a kind of linguistic time capsule. He had been born in Leicester in the mid-eighteen thirties and had first come to New Zealand in 1856. He returned to Britain for four years to manage a family business in Nottingham, before settling permanently in Dunedin in 1866. He was thus upholding standards of forty years earlier – and it is, of course, noteworthy that he made no mention of any version of [hivven]. Burton was an imposing figure whom contemporaries recalled striding importantly along the city's main thoroughfare, Princes Street, and his self-confidence is evident in the assertive quality of his letter to the local newspaper. But, in fairness to the man, it must be acknowledged that, as Dunedin's principal teacher of elocution – and founder of its Shakespeare Club – he could hardly fail to enter the debate.[94]

However, there are indications that Alfred Burton's defence of traditional vocal verities at the ends of the earth was being overtaken by pronunciation change back in Britain. The relatively frequent pairing of "heaven" with "seven" in Punch suggests that both words had become unambiguously bisyllabic. There is confirmation in a bizarre verse published in 1915, mocking the fate of ten members of the Territorial Army, the military reserve, who had been posted to India. It was based on a nursery rhyme which jauntily chronicled the tragic fate of ten youngsters, ending "and then there were none", a ditty that was unusual in that each rhyming couplet was effectively paired into a single line. (Its modern title, Ten Little Soldier Boys, effaces an earlier name that is no longer considered acceptable.) After one soldier dies of dysentery, another succumbs to snake-bite. "Eight Territorials hoped he'd rest in heaven; One took his topee off, and then there were seven." It was a curiously tasteless squib to publish in wartime, but its value here is that it was based upon a well-known underlying patter rhythm which confirms that "heaven" was intended as two full syllables.[95]

By the nineteen-thirties, the issue may be said to have been definitely resolved: heaven was bisyllabic. Although it was a pastiche of Portia's "Quality of mercy" speech in The Merchant of Venice, the Punch celebration of motor transport in 1936 hardly aspired to the status of immortal poetry: it rhymed "Shelley" with "smelly", but its culminating emphasis upon "Devon" offers a persuasive clue to the pronunciation of our key word. "The quality of petrol can be strained, / It springeth like a sudden lark to heaven / Out of the earth beneath, and it has gained / Power to pull charabancs from here to Devon."[96] Later that year, Punch called upon the same pairing for a theme that was also hardly Wordsworthian: an albino pied wagtail had been spotted at a sewage farm in Liverpool. "Were I a water-wagtail pied, / And free of all the countryside, / With feet for earth and wings for heaven, / I'd go to Cornwall or to Devon."[97] Of course, the possibility remains that "heaven" was still pronounced in clipped form, as is recommended by the Cambridge Pronouncing Dictionary to this day, but this would imply that "Devon" was similarly truncated, which seems unlikely. In 1938, Henry Martin specifically identified both "heaven" and "seven", the word so often paired with it, as double rhymes, with each syllable counting. However, he characterised them as 'feminine' rhymes, a term that strictly speaking describes a stressed syllable followed by one that is unstressed.[98] Thus to be sure that [hevv'n] had fully made the transition to [hevven], we need to find it rhymed with some word that was indisputably bisyllabic.

The linguistic smoking gun emerged from the politics of wartime Britain. While Churchill headed a coalition ministry that embraced all three political parties, the parliamentary system required some kind of opposition that could subject the government to scrutiny. The vacuum was filled by an ad hoc group of dissident MPs, loosely led by the Socialist Emanuel Shinwell and the Conservative Earl Winterton: a popular stage play inspired their joint nickname, "Arsenic and Old Lace". [99] The element of internal contradiction in the phrase 'loyal opposition' was particularly acute in a time of national peril: critics of Churchill's government had to tread carefully to avoid providing material for Nazi propaganda. Some felt that no such inhibitions restrained another freelance member of the opposition, an outspoken Welsh Labour MP. In 1942, this maverick was reproved in verse by A.A. Milne, the creator of Winnie the Pooh: "Goebbels, though not religious, must thank Heaven / For dropping in his lap Aneurin Bevan..." Great moments of national crisis sometimes produce enduring works of literature. This was not one of them – not even outraged patriotism can justify rhyming "Hitler" with "belittler" – but Milne's doggerel is important in pairing "heaven" with a surname that was incontestably bisyllabic.[100] One further example, already mentioned above, may also be cited here. In 1974, John Betjeman contributed a poem in celebration of the Warwickshire Avon. He traced the river's course downstream until the point "…where the tower of Tewkesbury soars to heaven / Our homely Avon joins the haughty Severn." Since he declaimed the verse at the opening ceremony (in the presence of the Queen Mother, no less), Betjeman presumably assumed that the rhyme would sit easily on the ear. Thus the poet who was widely regarded as Britain's last Victorian definitively endorsed the bisyllabic pronunciation of "heaven".[101]  

Some tentative conclusions Ideally, the conclusions to an essay should at least sound conclusive. Unfortunately, even after an enjoyable ramble through a random selection of the possible sources, I cannot project an illusion of oracular confidence. Instead, I offer five outline reflections, enough perhaps for others to take the enquiry forward.

First, we simply do not know and cannot know how words were pronounced before the end of the nineteenth century because sound recording had not been invented. (I do not know if there is a glossary of phonograph material, so I cannot say when "heaven" was first cut into a cylinder or disc.)

Second, balance of probability points to [hevven] as the formal pronunciation, certainly so accepted in the late nineteenth century, and predominantly recognised as such perhaps as early as the late eighteenth. That balance is admittedly slight. If the question is asked, "recognised by whom?", the answer lies in the compilers of pronouncing dictionaries, and  we should not forget that these enthusiasts had a very specific agenda that was focused upon reform rather than of reporting. They key evidence comes from Isaac Pitman, who at the age of 74, selected "heaven" in 1887 as one of the first two words whose spelling might appropriately and uncontroversially be adjusted in phonetic reform, and from Lord Grimthorpe, who was 78 in 1894 when he condemned those who rhymed "heaven" with "driven" and "riven". Grimthorpe, who was habitually dogmatic, might have chosen to forget variant versions heard in early life, but Pitman would have hardly have committed himself so strongly to a sample term had his proposal been open to the objection that there was no agreed phonetic interpretation of the word.

Third, nonetheless, there is too much evidence from hymns and other verse to dismiss the alternative altogether. It is difficult not to believe that churchgoers and lovers of poetry were comfortable with rhymes that pointed to a spoken form as [hivven], especially in the eighteenth century. It is clear that there was a great deal of overlap between vowel sounds: this is particularly true of words spelt with a short –o– (or so we would expect from wider usage) but which are pronounced with a short –u–, of which around fifty survive, employed comfortably in company, causing no worry either to mothers or brothers. We are less familiar today with words spelt with a short –e– but pronounced with a short –i–, but "pretty", "England" and "English" presumably draw upon some older tradition. Walker disapproved of [yit] for "yet" and poured scorn upon those who called an engine an [injin], but even he accepted [yiss] for "yes", with the significant comment that the vowel had been "worn into a somewhat slenderer sound than what is authorised by the orthography; but e and i are frequently interchangeable". This points to a two-tier system of pronunciation – an official version but a generally tolerated popular adaptation. The distinction was neatly defined by Punch in 1842, when it noted the Cockney pronunciation of the London street, St Mary Axe, which was generally known as "Simmary". This telescoped form was "not the real, and, so to say, organic pronunciation of the term but rather the synthetic and popular one".[102] In the same spirit, it seems plausible to postulate that [hivven] was a widely used if semi-underground form of "heaven". The key point here is acceptance of widespread imprecision in the interpretation of vowels, which even the purist Walker could not deny. Educated British observers tend to regard the telescoping of short –e– and short –i– words that is often encountered in Australian and New Zealand English as an antipodean corruption, but it may well be attributed to forms of speech imported in the early years of nineteenth-century colonisation.

Fourth, in attempting to recapture its past pronunciation, it is important to note that, until well into the nineteenth century, "heaven" had two unusual features. First, it was neither one syllable or two, but (whether voiced as [hevv'n] or [hivv'n]) occupied a curiously fluid midway position. Second, it was not some workaday word referring to amenities or tools, but a term that commanded a reverence that may now seem almost superstitious. The combination of structure and aura placed considerable limitations upon the use of "heaven" in poetry and hymns.  The selection of acceptable rhyming words was limited, first, to those that also had a truncated second syllable and, second, to concepts that conveyed respectfully religious sentiments. These restrictions resulted in an extremely narrow range of acceptable terms for poetic partnership. Since "leaven" and "seven" were secular and frivolous, while "given" and "forgiven" were capable of conveying devout sentiments, hymn-writers (and, to some extent, poets too) took advantage of the postulated two-tier system of dictionary and demotic pronunciations and placed the emphasis of their verse upon [hivven] rather than [heaven]. However, by the end of the nineteenth century, these innate peculiarities of syllables and sanctity were in retreat. Here, I can assert only that the transitions took place, while offering far less confident analysis of the process.

It seems reasonable to conclude that, for most people, "heaven" definitely became unambiguously bisyllabic, thereby broadening the range of potential rhyming words with which it could be paired. Conformity of the spoken language with the increasingly widely comprehended written form also reinforced the definitive shift from [hivven] to [hevven]. The most likely explanation for these developments would seem to lie in the growth of popular literacy, but precisely when this happened or how the change took hold are questions that remain shrouded in mystery. It is difficult to identify any process by which younger members of the lower social classes could have played the role that Smart had attributed to "the smaller literati". The explosion of the press in the mid-Victorian period certainly points to a growth in popular literacy, but it is unlikely that the masses read publications that had much to say about Heaven. All we can say is that a bisyllabic "heaven" was condemned as "vulgar and childish" in 1791, but seems to have been generally accepted one hundred years later and was formally endorsed in an English textbook designed for India half a century beyond that. The second change saw "heaven" shed at least some of its taboo quality, its hands-off status as a word to be treated reverently and in a sacred context. Here, again, it is easier to trace the outcome than to identify the process that brought it about. There was a Victorian theological debate on the nature of Heaven, which predominantly moved away from the notion of a physical, if celestial, location after the afterlife in favour of broader and more spiritual concepts. Yet it is far from clear that these intellectual constructs exercised any widespread influence. Nor should we forget the counter-currents that responded to the challenge of changing times by seeking a return to antique certainties. From the eighteen-forties, there was a small but noteworthy trickle of defections from Anglicanism to the embrace of Rome, attracted by the conservative theology of the Catholic Church. At the opposite end of the social scale and the far end of the sectarian spectrum, the eighteen-eighties would see the launching of the Salvation Army, with its Protestant fundamentalist beliefs. Yet, once again, we can point to the outcome, even if we cannot isolate its causes: in the columns of Punch, at least, "heaven" came to be treated in a much more light-hearted, even friendly manner, applied to the batting of W.G. Grace, the port served at  the Inns of Court, hedonistically rhymed with "seven", "Devon" and eventually paired with the notably heterodox figure of Aneurin Bevan. 

Fifth, it seems worth defining, even confessing the limitations in the research underpinning this reconnaissance. For me, at least, it has been an enjoyable journey into unforeseen byways, but any illusion of breadth should not be permitted to disguise its relative lack of depth. It must be faced that I have only traced "hivven" in printed form in attempts to render Irish brogue (and some Scots vernacular) in amateur phonetics around 1900. This undoubtedly represents a lacuna in the material considered. It may represent evidence that a once-widespread pronunciation had retreated to the Celtic fringe where metropolitan readers might dismiss it as risible. But it might equally be taken as a weakness in the overall argument. Hence I offer the suggestion here that there may well be some contemporary source that did indeed parody the spoken language of clergy, choirs and congregations: perhaps a Victorian novel, a gossipy memoir or a scornful secularist tract, lurking on library shelves or in online scans for some future PhD student to discover. In passing the baton to others, I reiterate my opening hypothesis. The period from late eighteenth century to the mid-nineteenth was a golden age both for Protestant hymnary and for Romantic poetry. In that era, it was not unusual for words to exist in two forms, one approved or dictated by that new form of linguistic policing, the dictionary, the other employed in common speech that elided, coalesced and confused vowel sounds – in this case, [hevv'n] and [hivv'n]. Constraints of metre and fear of irreverence led most versifiers in those prolific years to opt for those rhyming words for "heaven", especially "given" and "forgiven", which satisfied both criteria while sitting comfortably with the popular spoken form. Unfortunately, the spread of relatively low-level literacy in Victorian times hastened the eclipse of the demotic rendering in favour of a read-and-say version that reflected the word in its printed form. Perversely, this linguistic change coincided with the creation, in 1861 and 1870, of the two great Anglican hymnbooks which, in the case of rhymes for "heaven", fossilised a usage that was increasingly alien to those who participated in religious worship. Hence, my own puzzlement, as I gave voice in a spiritually arid school assembly a century later, at the pairing of "Praise, my Soul, the King of Heaven" to that overwhelming but jarring phrase, "Ransomed, healed, restored, forgiven".

ENDNOTES  I owe particular thanks to Trevor Harrison and to Andrew Jones. Editors of parish magazines and newsletters are welcome to reuse this material, but are asked to cite the url: https://www.gedmartin.net/martinalia-mainmenu-3/415-praise-my-soul-hivven.  Martinalia also includes ""Housen" – evidence for the survival and decline of an Essex dialect plural": https://www.gedmartin.net/martinalia-mainmenu-3/222-housen-evidence-for-the-survival-and-decline-of-an-essex-dialect-plural

[1] In 1834, Christopher Wordsworth, Master of Trinity College Cambridge, defended the imposition of Chapel attendance upon undergraduates by explaining that the alternative lay "between compulsory religion and no religion at all". A reformist Fellow, Connop Thirlwall, commented: "I cannot indeed draw such delicate distinctions". This was my experience of school religion.

[2] I make no apology for conflating the words by H.F. Lyte, published in 1834, with the tune by John Goss, composed in 1868: for a century and a half, the two have been inseparable. There have been some attempts to amend Lyte's text. Methodists had some problems with the role of angels in verse 4. In verse 2, Unitarians (and some trendy Anglican clergy) have changed "fathers" to "forebears", vandalism condemned by a correspondent of The Times (9 May 2001) as "arrogance to the hymn-writer" and "condescension to the congregation". The fourth line of the first  verse is sometimes reworded in less convoluted versions, while "thy" is rendered as the more conversational "your". In my schooldays, the chorus, "Praise him! Praise him!...", as labouringly sung in London accents by 600 unmusical adolescents, evoked the image of a lorry [now called a truck] struggling up a hill: in postwar Britain, commercial vehicles were lumbering machines. Many versions substitute the cheerier "Alleluia!". However, there is not much that can be done about the resounding celestial monarchism of the first verse, and the divine pronouns are definitely He / Him. Praise, My Soul, the King of Heaven is perennially popular. In 1936, the Bishop of Chelmsford sought, not very seriously, to place it on a list of 9 overworked hymns to be subject to a 12-month ban. Travelling round his diocese on Sundays, he frequently found himself singing it twice and, on one occasion, three times. In 1995, a poll among listeners to BBC Radio 4 placed Praise, My Soul, the King of Heaven 7th in its 'hymn parade' (Dear Lord and Father of Mankind came first, Abide With Me was 4th). It was sung at the funeral of President George Bush in 2018, and the Coronation of King Charles III in 2023. According to the Hymnary website, it appears in 537 hymnbooks. There is a useful note by J.R. Watson in The Canterbury Dictionary of Hymnology: http://www.hymnology.co.uk/p/praise,-my-soul,-the-king-of-heaven. In his 1999 study, The English Hymn..., Watson referred to "Lyte's hymnic art, which joyfully exploits patterns, rhymes, rhetorical flourishes, to make strong lines and verses leading, usually, to a powerful conclusion" (ch. 13). His splendid phrase, "the cable‐stitch of Lyte's verbal pullover", certainly captures "Ransomed, healed, restored, forgiven". 

[3] I am encouraged in this by the verdict of Joseph Wright, editor of the English Dialect Dictionary (cited below, i, xvii) in 1898: "An elaborate transcription is useless to people who have not had a practical training in phonetics."

[4] To use the term 'methodology' to characterise the preparation of this study may seem exaggerated. Briefly, I begin by establishing that Lyte's use of [hivven] was typical of contemporary hymn-writers. There follows an outline survey of versification employed by a random selection of English poets in the two centuries before Praise, My Soul, the King of Heaven. This was followed by a series of excursions into a later period, using newspaper sources and, in particular, popular verse, especially from the humorous magazine, Punch. There are at least three shortcomings in the approach adopted. First, there is almost certainly far more relevant material than is touched upon here, which may well include explicit contemporary discussions of pronunciation. Second, my investigations are based upon poetry websites and scanned volumes of collected works, mostly through the Internet Archive (archive.org). Since the quotations are easily traceable, I have been sparing in the use of endnote references. Search commands depend upon optical character recognition systems, miraculous technology but still with understandable limitations: it cannot always detect words that are hyphenated or abbreviated with an apostrophe (as was often the case with "heav'n"), nor is it comfortable with some older typefaces. Third, my own numeracy is not entirely reliable, even in the single-digit range, and such statistics as are cited here should be regarded as indicative rather than precise.

[5] A similar disagreement within source materials briefly enlivened rural Wales in 1888. A Mr J.P. Howell, a local solicitor and prominent citizen of Cardigan, argued that "again" should be pronounced [agen]. He was handled somewhat roughly by a correspondent who borrowed the name of a Victorian comic character, 'Dundreary'. Dundreary asked " how it is possible to arrive at the sound – the pronunciation – save by the spelling, or through the medium of poetry?" A "hasty scan" of Shakespeare established that "he rhymes 'again' thirteen times with words ending in 'ain', and but once with 'en'." Dundreary found it "inconceivable that Shakespeare, avoiding the true sound of the word ..., rhymes it falsely 13 times, and truly but once and that once, by the way, among the loosest, admittedly, of his verses." In reply, Howell challenged his assailant to explain "why is it that he is so chary and shy of dictionaries; that should be to him as the storehouse and mirror of the sound." Although Dundreary cited other poets, he was perhaps on weak ground in appealing to Shakespeare, whose enthusiasm for eye rhymes made him an unreliable source. And Howell was right to appeal to the dictionaries. Walker's Critical Pronouncing Dictionary of 1791 [cited below], firmly declared: "Again and against sound as if written agen and agenst." Cardigan Observer, 1, 8 October 1887. The issue arose again in 1933, on this occasion in the columns of The Times. A correspondent conceded that "with poetic licence, in some of our hymns the word 'again' is made to rhyme with 'main'", but appealed to Isaac Watts  who had written: 'Angels descend with songs again, And earth repeats the loud Amen.'" The couplet came from a hymn of 1719, Jesus Shall Reign Where'er the Sun. The Times, 27 September 1933.

[6] Edison's phonograph was invented in 1877 but not marketed on a large scale for another decade. Sound recording was not widely used before the 20th century. For the difficulty in identifying even the accent of a public figure, see Ged Martin, "Sir John Eh? Macdonald: recovering a voice from History": https://www.gedmartin.net/published-work-mainmenu-11/15-sir-john-eh-macdonald.

[7] "Though it cannot be denied, that there is still some difference in our orthography, owing to the affectation of writers and caprice of printers; yet it is observable, that, since the publication of Dr. Johnson's Dictionary; and particularly the last edition, less variation is found." A Vocabulary of Such Words in the English Language as are of Dubious or Unsettled Accentuation (London, 1797), i.

[8] R.W. Chapman, ed., Boswell: Life of Johnson (Oxford, 3rd ed., 1953), 470. Walker in 1791 [cited below] commented (lxiv): "The word great is sometimes pronounced as if written greet, generally by people of education, and almost universally in Ireland; but this is contrary to the fixed and settled practice in England." Modern usage would suggest that he may have been misinformed about Irish vernacular speech. 

[9] Confusingly, Sheridan prescribed two opening syllables for the subject, "geography".

[10] The Oxford English Dictionary  noted that [bum-] was "formerly usual" in the British Army, and accepted Walker's version as variant for "bombard" and "bombardier".

[11] The pronunciation of –o– words as short –u– is deeply embedded in contemporary English, but some eighteenth-century variants have since fallen into line with their spellings. Walker listed many examples still so pronounced today: above, affront, among, amongst, borough, brother, colour, come, comely, comfort, comfrey, company, compass, conjure, constable, covenant, cover, covet, covey, done, dove, dozen,  front, glove, govern, honey, love, Monday, money, mongrel, monk, monkey, month, mother, none, nothing, once, one, onion, other, oven, shove, shovel, smother, some, Somerset, son, sponge, stomach, thorough, ton, tongue, wonder, worry. In one case, "romage" (the noun and verb referring to energetic searching), the modern spelling has become adapted to the pronunciation ("rummage"), although Walker was faithful to the usage of Jonathan Swift. He also included "pommel", both as a noun referring to a knob on a saddle or the hilt of a sword, and as a verb meaning "to punch". The noun is generally now pronounced as spelt, the verb has adjusted to "pummel". In other instances cited by Walker, the written form has eliminated now-forgotten short –u– speech forms: borage, comrade, combat, conduit, coney, dromedary, pomegranate, sovereign. Officially, "covert" retains the short –u–, but is often pronounced with a long –o–, especially when linked to "surveillance". Other words cited by Walker are now rarely used, if at all: cozen, discomfit, doth and dost (past tenses of do), pother. The noun "sloven" has largely disappeared  but the adjective "slovenly" retains the short –u–. Walker also listed examples where –or– had become –ur–: word, world, worse, worship, worth. It may be noted that "dove", "glove" and "love" have resisted the tendency to pronounce words as they are spelt, the final –e– usually being taken as indicating a long vowel. Walker refused to follow Sheridan in approving a short –u– in command, commit, commence. He recommended a short –o– in compare, compel, commence, compose, "though I am sensible that, in colloquial pronunciation, they all approach nearer to the short u". J. Walker, A Critical Pronouncing Dictionary and Expositor of the English Language... (London, 1791), esp. 21-2. Walker's introductory comments were paginated; entries in the dictionary section can be traced alphabetically.

[12] No doubt as a Londoner, I still colloquially conform to Walker's usage, contrasting "where's me cap?" with "I can see your phone but I can't find my watch", but it is startling to discover that it was once regarded as a rule of accepted speech.

[13] This can still be the case with enthusiasts for correct pronunciation. See Ged Martin, "Pronunciation for presentation: effective English for presentations. Notes for students (and everyone else)": https://www.gedmartin.net/using-english-effectively/167-pronunciation-for-presentation.

[14] Lord Chesterfield's Letters to His Son ... (London, 1774), Letter LXXXIII.

[15] Punch, xxvii, 1854, 7.  Early volumes of Punch in library collections are generally reprinted editions that omitted the dates of individual issues.

[16] Walker, Critical Pronouncing Dictionary, 13.

[17] The Oxford English Dictionary traces "Varsity" to the 1840s. It is possible that it was coined in response to the emergence, between 1826 and 1836, of the upstart University of London.

[18] I take as a random example, "a Ginral Vestre hild by the Prishnors" at Corringham in Essex in 1776: L. Thompson, The Story of the Land that Fanns... (Chelmsford, 1957), 18. Walker would have preferred to sound each syllable in "nominative", but felt unable to resist popular error. "This word, in the hurry of school pronunciation, is always heard in three syllables, as if written Nomnative; and this pronunciation has so generally prevailed, that making the word consist of four syllables would be stiff and pedantic." My own recollection is that twentieth-century usage in Britain had no such problem in voicing the word in full. In the twenty-first century, it is doubtful whether inflected languages are taught on any large scale, and the question probably does not arise.

[19] B.H. Smart, Walker Remodelled... (London, 1836), xxxiv, xl; Walker, Critical Pronouncing Dictionary, xii.

[20] Punch, v, 1843, 7. Punch (iv, 169-70) defended Cockney against "the pronouncing dictionary of Valker".

[21] He added that "while persons so endowed are to be found ready to undertake the employment, (and they always will be found,) the rest of the world will be content to follow their track, satisfied if clearly understood, and that their manner neither prejudices their matter by uncouthness, nor calls off attention from it by obtrusive nicety or unusual preciseness. But, while it becomes every sensible speaker to adopt all changes for the better, as soon as he safely may, it equally becomes him to oppose such as have no recommendation but caprice and fashion, and which would injure instead of improve the audible structure of our language." Smart, Walker Remodelled, xl.

[22] G.H. Powell, ed., Reminiscences and Table-talk of Samuel Rogers...  (London, 1903), 193. The earliest rustic "Lunnon" that I have noticed was in Punch, 16 April 1881, 187, but there may well be earlier examples. The affected speech of the British elite may seem objectionable, but it was at least more accessible to the masses than was the case in Prussia and Russia, where aristocrats expressed themselves in French, and often could not speak the local peasant tongue at all.

[23] W. Ward, William George Ward and the Oxford Movement (London, 1889), 116-17.

[24] This particular report was published in the Rhyl Advertiser, 30 April 1887 (via the National Library of Wales Welsh Newspapers online archive) but I suspect that it had appeared elsewhere. There is an inverted example of the relationship between [dome] and [doom]: William the Conqueror's Great Survey of 1086 is known as  "[Doomsday] Book", and is sometimes so spelt, but in scholarly texts it is "Domesday".

[25] Powell, ed., Reminiscences and Table-talk of Samuel Rogers, 193. Rogers also complained about the emphasis upon the final syllable in "contemplate".

[26] J.C. Beal, English Pronunciation in the Eighteenth Century... (Oxford, 1991), 37. Herrick in 1648 wrote "but all things even / Make for thy peace, and pace to heaven".

[27] O. Chadwick, The Victorian Church, ii (2nd ed., London, 1972, cf. 1st ed., 1970), 397-9. Praise, My Soul, the King of Heaven appeared in both collections. Bickersteth used the tune by Goss. In his 1984 discussion, "What Makes a Good Hymn Text?", T. Dudley-Smith argued that "a concluding rhyme gives shape and finish to a verse, gives that satisfying inevitability – without, one hopes, being too predictable – and helps to prevent the whole from becoming unstructured and over-facile": https://hymnsocietygbi.org.uk/wp-content/uploads/2017/11/T50-What-Makes-a-Good-Hymn-Text.pdf. It seems fair to assume that this had been the underlying principle of hymn-writing throughout the previous two centuries. 

[28] Catherine Winkworth translated other hymns: in Christ the Lord Is Risen Again!, she rhymed "heav'n" with "forgiv'n". In 1844, as a 16-year-old schoolgirl, she is said to have invented one of the few Victorian jokes that have stood the test of time. Sir Charles Napier, commander of the Bombay Army, was ordered to suppress a rebellion in the province of Sindh, which was not then under British rule. Napier sensationally exceeded his instructions and annexed the territory outright. Winkworth who (unusually for a girl) was studying Latin, suggested that Napier should have sent a single-word report to London, "Peccavi" (I have sinned). The punning witticism was submitted to Punch, which published it on 18 May 1844. Watson, The English Hymn, has a chapter on the female hymnists.

[29] I have had less success in tracking down English Catholic hymns. An American publication (Catholic Hymns, Albany, NY, 1860) has four rhymes of "Heaven" with "given". Before Vatican II, some of the most popular Catholic hymns were sung in Latin, e.g. Adeste Fideles for O Come All Ye Faithful.

[30] D. Crystal, The Cambridge Encyclopedia of the English Language (Cambridge, 1995), 55.

[31] However, another London-area place-name, Chelsea, provides less persuasive evidence. The Place Names of Middlesex records only a handful of examples of the first syllable of Chelsea being recorded as "Chil-", and all but one of these occurred between 1042 and 1319 – well before the time of Gower. The outlier, "Chilcy", dates from 1655. Perhaps the comparison proves that attitudes to speech have always been more casually fluid on the east side of London than on the west.  J.E.B. Gover et al., The Place Names of Middlesex (Cambridge, 1942), 148-9, 85-6.

[32] Martin Benson was Bishop of Gloucester.

[33] Westcott was not a recognised poet, but wrote some verse during his days as a student at Cambridge. A.W. Westcott, Life and Letters of Brooke Foss Westcott ... (2 vols, London, 1903), i, 133. It is only fair to add that his What is Life? also contains "A glimpse of heaven in rock and wood, / In rivulet and torrent flood, / In temple, tomb, and solitude – / Oh, such is life." Even if allowance is made for a Birmingham accent, this does not make him a reliable source for accurate rhyme.

[34] Walker, Critical Pronouncing Dictionary, 14, 297. 

[35] Chapman, ed, Boswell: Life of Johnson, 470.

[36] Beal, English Pronunciation in the Eighteenth Century, 58. In his entry on "fabrick", Walker noted that the letter –a– in the first syllable "seems floating between long and short quantity": he opted for the short form, giving various reasons for his verdict.

[37] In Adeline (1830), Tennyson produced a couplet: "Thou that faintly smilest still, / As a Naiad in a well". A thoughtful review of the Poet Laureate's career in 1869 objected to "the rude blemish of such a rhyme... in a laboriously polished verse". But it is equally possible that, if the young Tennyson did indeed take so much care over Adeline, he assumed that readers would interpret "well" as [will]. Four decades later, any such usage was forgotten. The Times, 31 December 1869.

[38] Punch in 1844 commented upon two other vowel shifts that were common but rarely noted, and associated with specific occupations: London cabmen touted for trade crying "keb, sir!", while barristers addressed judges as "m'lud". The latter was hardly the product of illiteracy, but perhaps owed its currency to the need to distinguish between "his lordship" on the bench and "my lord" in the pages of Burke. (Punch, vii, 1844, 99)

[39] Walker did not comment on "get". In T. Sokoll, ed., Essex Pauper Letters 1731-1837 (Oxford, 2001), there are six spellings as "git" between 1825 and 1834, plus a "giting" (getting) in 1836. There are numerous other –i– for –e– spellings in the 736 begging letters. Regrettably, none of them invoked the blessings of Heaven upon parsimonious parish officials. E. Gepp, An Essex Dialect Dictionary (London, 1923), 149, noted: "Short i takes the place of e in git (get), hin (hen), and 'it (yet)". Walker's hostility to [yit] for "yet" suggests that these usages were not confined to Essex.

[40] For instance, Walker simply rendered "girl" as "gerl". But Punch in 1843 noted that "gal" was "[t]he common person's pronunciation of a young maiden" (v, 1843, 271).  A New Zealand newspaper commented in 1932: "The word 'girl' is, perhaps, unique in English from the point of view of pronunciation. While men rhyme it with 'pearl', women give it a sound which degenerates into 'gal', while some patrician dames pronounce it 'gell'." (Christchurch) Press, 18 November 1932.

[41] In 1769, Aveley in Essex recorded a payment "for Licker for the Men that brought the Ingen to the fire". Thompson, The Story of the Land that Fanns, 19.

[42] A century later, an authoritative American handbook on pulpit oratory, A Treatise on the Preparation and Delivery of Sermons, baldly advised preachers to be cautious in any experimentation with words. "Never depart from the pronunciation of them which is common among educated people, unless there is something real to be gained by it." First published by J.A. Broadus in 1870, this quotation comes from a 23rd edition, revised by E.C. Dargan, and published in New York in 1898, 518n. Of course, the statement would only be helpful in this context if we knew how educated people pronounced "heaven". J.J. Blunt, The Acquirements ... of the Parish Priest, (2nd ed., London, 1857, cf. 1st ed. 1856), 8-9 merely advised that "as a general rule, words of Saxon origin are more safe than those of Norman or Latin derivation; they, at least, will be intelligible, both to high and low, and will commonly be as capable of sustaining an argument from the pulpit with dignity as any others." I am not persuaded by the report of a Welsh preacher who startled his congregation with the statement that "the Devil was not a curate". Allegedly, he was accusing Satan of inaccuracy, surely one of Beelzebub's lesser failings. However, in the first half of the 19th century, Welsh clergy were notoriously badly educated. Cambrian News, 9 March 1894.

[43] The couplet comes from An Essay on Man: Epistle I.

[44] Smart, Walker Remodelled, 200, 694, 1, xxi. Gepp, Essex Dialect Dictionary, 149, noted that "deaf" was pronounced [diff]. Until about 1900, "cross" and "loss" were pronounced to rhyme with "force" but, in England, the vowel sound was modified to rhyme with "moss". (R.W. Burchfield, ed., Fowler's Modern English Usage (3rd ed., Oxford 1996, cf. 1st ed. 1926), 630.) Similarly, in The Pirates of Penzance (1879), Major-General Stanley asks: "When you said 'orphan', did you mean 'orphan' – a person who has lost his parents, or 'often', frequently?" Ireland's state broadcaster, RTE, still favours the older version.  

[45] It was so declared in J.E. Worcester's A Dictionary of the English Language (Boston, 1860), 671.

[46] Quarterly Review, xix (1818), 206.

[47] Leigh Hunt had employed a similar rhyme in his jaunty and jokey The Feast of the Poets (1811), a revelry that brought together deities and immortals, and which was brought to a close by the departure of one of the former. "Thus chatting and singing they sat till eleven, / When Phoebus shook hands, and departed for heaven". The rhymes are brutal.

[48] M. Wheeler, Heaven, Hell, and the Victorians (Cambridge, 1994) is important here. Wheeler argued (120-1) that, until the 18th century, Protestant conceptions of Heaven were "theocentric", i.e. focused on God ruling in majesty with Christ on his right hand. Thereafter, "anthropocentric" models became more popular. These emphasised Heaven as a meeting place, especially for reunions of married couples. Wheeler caustically described such visions as "more like a middle-class suburb in the sky than the City of God". Praise, My Soul, the King of Heaven was unambiguously theocentric. In a sermon in Westminster Abbey in 1877, F.W. Farrar rejected the notion of Heaven as a "golden city in the far-off blue", arguing instead that it was "a state rather than a place …. a temper rather than a habitation". Farrar's theology was controversial, but his redefinition of Heaven was part of the more relaxed atmosphere of the times. A third of a century later, in 1914, there seems to have been little comment when W.R. Inge, the habitually blunt Dean of St Paul's, virtually gave up on the concept altogether. He recognised "that uneducated people did suppose the teaching of the Church to be that Heaven was a literal place where God and the Angels lived, and where good people went when they died or after the Day of Judgment…. But the average man … knew enough astronomy to feel the absurdity of placing it either inside or outside the solar system. So[,] many of the clergy were perplexed themselves, and said as little about Heaven as they decently could." Inge would subsequently argue that it was not 19th-century rationalism that had destroyed the idea of Heaven as a physical location, but the discoveries of Copernicus and Galileo three centuries earlier. "If the earth is a planet revolving round the sun, and if the solar system is only a speck in infinite space, the old geographical heaven and hell must be abandoned." This retreat opened the way to thinking about Heaven in less absolute terms. "What we call heaven and hell are not two places; they are the two ends of a ladder of values. We shall all stand somewhere on the ladder, where we deserve to be." F.W. Farrar, Eternal Hope… (London, 1878), 25, 19; The Times, 22 May 1914; W.R. Inge, Outspoken Essays: Second Series (London, 1922), 36, 24.

[49] There is, however, a notable absentee, at least from my relatively cursory surveys, a kindred participle that, at first sight, would seem firmly anchored in the realms of the devout. "Shriven" described worshippers who had secured absolution from their sins through Confession.  In terms of concentrated religiosity, it would seem to outscore the other four altogether, yet it never appeared in hymns and does not seem to have been prominent in verse. ("Shriven" appears in the headword index of neither Bartlett's Familiar Quotations nor the Oxford Dictionary of Quotations.) No doubt it was already becoming archaic by the nineteenth century, but an aura of medievalism alone would hardly have deterred Keats. The problem, of course was that it was a Roman Catholic concept, in conflict with a fundamental Protestant principle that entry into Heaven did not depend upon the intermediary role of any priesthood. Post-Reformation English literature was imbued with Protestantism and overwhelmingly created by writers who, at the very least, found it prudent to conform to the official culture. There were exceptions. Alexander Pope died in the faith of his parents, but essentially wrote for an audience that saw itself as rational. By the time the Jesuit Gerard Manley Hopkins was active, any possibility of a [hivven] pronunciation was long forgotten. It is perhaps curious that Newman did not use "shriven" in The Dream of Gerontius, a deathbed poem. (Nor did he employ it in his magisterial spiritual autobiography, the Apologia.) In birthday verses for his friend George Meredith in 1863, the diarist William Hardman rhymed the verb form, "shrive", with "thirty-five". The two called themselves Robin Hood and Friar Tuck and Hardman was here adopting the persona of a medieval mendicant. S.M. Ellis, ed., A Mid-Victorian Pepys … (London, 1923), 253. 

[50] I mention here an enigmatic formulation in the Authorised Version: "The kingdom of heaven is like unto leaven..." (Matthew  xiii, 33). Were the translators having fun?

[51] Hymns were also sung in the syncretist Labour Church of the 1890s. One contained the couplet: "I feel, though I be sacrificed, / That one must follow Buddha, Christ." K.S. Inglis, Churches and the Working Classes in Victorian England (London, 1963), 238. Christina Rossetti rhymed "Christ" with "sufficed" in From House to Home. Neither rhyme appears in Hymns Ancient and Modern. In the Church Times (4 June 2002), David Wilbourn asked: "Is God happy with being condemned to rhyme with sod for all England’s hymn-singing eternity?" Wilbourn also declared that "hymns are the oral version of stained-glass windows, their memorable and usually patronising rhymes aimed at instilling doctrine into the illiterate masses". (Quoted https://hymnsocietygbi.org.uk/wp-content/uploads/2017/11/T68-When-Words-Call-the-Tune.pdf) Even an atheist may roll his eyes at so bold a sentiment. 

[52] "Oh! pleasant exercise of hope and joy! / For mighty were the auxiliars which then stood / Upon our side, we who were strong in love! / Bliss was it in that dawn to be alive, / But to be young was very heaven – Oh! times, / In which the meagre, stale, forbidding ways / Of custom, law, and statute, took at once / The attraction of a country in romance!" (italics added) The French Revolution as It Appeared to Enthusiasts at Its Commencement (1804).

[53] Most volumes of Punch are available online via either the Internet Archive (archive.org) or the Hathi Trust: https://onlinebooks.library.upenn.edu/webbin/serial?id=punch.

[54] This particularly applies to the idiosyncrasies of surnames. For an example, see Ged Martin, "How to pronounce Parnell and say O'Shea": https://www.gedmartin.net/martinalia-mainmenu-3/371-pronounce-parnell.

[55] The Times, 11 April 1842. The mockery of Wakley was in the spirit of Hood, but the poem is not attributed to him. It carried the initials "P.S.", hardly a hint of authorship by Shelley, who had been dead for 20 years. The verse is incidentally useful in clarifying the pronunciation of Wakley's unusual surname, which was rhymed with "blackly" and "slackly".

[56] Punch, 8 October 1859, 143.

[57] Punch, 21 January 1860, 24. The poem concludes: "And all the galaxy that fused / Thelr splendours into one, / When William ceased, and Anne begun / And state-craft writer-craft abused. / Who knew and treasured of all these / What was worth treasuring, more than he / Who to their silent company / Has last gone down, from life and ease?"

[58] Punch, 23 December 1871. Later attributions of the word to Irish speech were more likely to be rendered as "hivven".

[59] The versifier expressed concern at the revolutionary movement in Paris following the defeat by Prussia: "To make low high, high low, rough places even, / With fist and fire to enforce love on humanity, / The Commune may be led by light from Heaven – / But on the whole I prefer Christianity." Punch, 8 July 1871. The first line was probably a loose recollection of Isaiah, chapter 40, via Handel's Messiah. It is not great poetry.

[60] Punch, 27 January 1872, 41.

[61] Punch, 16 January 1869, 16. J. Wright, ed., The English Dialect Dictionary..., iii (Oxford, 1905), 123 noted that "heaven" was also spelt "heeaven" in the North Riding of Yorkshire, although only one example was cited. Wright commented that "it is sometimes found extremely difficult to ascertain the exact pronunciation and the various shades of meanings, especially of words which occur both in the literary language and in the dialects", and offered no detail about "heaven". The English Dialect Dictionary..., i (Oxford, 1898), v.

[62] E.g. Punch, 8 December 1888, 270; 2 March 1889, 102; 25 May 1889, 248; 15 March 1890, 123 ("driven"); 30 March 1895, 197.

[63] E.g. Punch, 1 December 1883, 264; 29 June 1889, 312; 21 September 1889, 138.

[64] Punch, 3 November 1888, 204.

[65] Punch, 30 August 1890, 102.

[66] Punch, 25 May 1895, 241.

[67] The Times, 29 September 1887. Pitman's proposal may have triggered an underground joke which surfaced in Canada forty years later. It concerned a man "who in his philological innocence ... hoped the reformers would leave 'heaven' the way the apostles wrote it". Globe (Toronto), 8 November 1927. But, soon after, a Cambridge Professor of Divinity, William Emery Barnes, singled out the word as an offence against common sense: "how seldom it is that we can appeal to the child's ear, when the child makes a mistake in spelling! Spell 'chest.' 'Good child!' Now spell 'breast – head – heart – heaven!'" The Times, 24 September 1934.

[68]  Punch still occasionally published verse that rhymed "heaven" with "given", but increasingly rarely. Australian and New Zealand newspapers (consulted via Trove and PapersPast) published some stories in Irish and Scots dialect, using "hivven", an indication of the retreat of a once-acceptable pronunciation to the provincial fringe. Examples attributed to the Irish include (from Australia): Daily News (Perth), 12 February 1902 (plagiarising the Pall Mall Gazette), Muswellbrook Chronicle, 19 April 1902, Brisbane Courier, 16 April 1924, Bunbury Herald, 12 November 1926; (from New Zealand), Ashburton Guardian, 13 January 1910, (Christchurch) Star, 22 July 1914 (taken from the Cork Chronicle, a Unionist newspaper) and Mataura Ensign, 19 July 1916 ("Comical Loot Stories" from the Easter Rising). Attributed to Scotland were alleged examples of 'hivven' in kailyard dialogue in Marlborough Express (New Zealand), 18 March 1895, Western Mail (Perth), 10 December 1897 and McIvor Times (Victoria), 14 May 1900. There are many more such examples, but they are tedious and irritating. See also Cambrian News (Aberystwyth), 28 June 1901, for "hivven" as an Irish pronunciation (consulted via National Library of Wales online newspaper archive).

[69] The Times, 25 August 1894.

[70] The Times, 30 August 1894. It is worth noting that J.J. Blunt, in his 1856 lectures, The Acquirements ... of the Parish Priest, made only two passing references to hymns, and evidently did not see the provision of music or the organisation of a choir as part of an incumbent's responsibilities.

[71] The Times, 27 August 1894.

[72] The Times, 28 August 1894. It was perhaps indicative of his contra mundum attitude that, on being raised to the peerage in 1886, Edmund Beckett should have taken his title from one of Yorkshire's less felicitously named villages. He was the principal ecclesiastical court judge in the Province of York, serving as Chancellor and Vicar-General from 1877. At his death in 1905, he left £1.52 million (worth about £230 million in 2024), a fortune that had had enabled him to pay for the restoration of St Albans Abbey from his own pocket. He dogmatically insisted on imposing his own ideas of Gothic architecture on the building, which led to an attempt to coin a verb "to grimthorpe", meaning to botch the restoration of a historic monument. He was also a mechanical genius, and designed the clock machinery for the Elizabeth Tower (popularly known as "Big Ben").

[73] H.W. Fowler, ed. E. Gowers, A Dictionary of Modern English Usage (rev. ed., Oxford, 1965, cf. 1st ed., 1906), 483.

[74] D. Vincent, "The invention of counting: the statistical measurement of literacy in nineteenth-century England", Comparative Education, l, 2014, 266-81; J.F.C. Harrison, The Early Victorians 1832-1851 (St Albans, 1973 ed., cf. 1st ed,, London, 1971), 164. Some early-19th schools taught reading but not the potentially seditious skill of writing. Few children remained at school beyond the age of 11, in most cases a decade or longer before they married. Lack of opportunity to wield a pen may have cost many their basic ability to write. In Wales, literacy was one of the weapons of anglicisation, and percentages signing marriage registers were notably lower. Scotland did not count until 1861, when over seven-eighths of men and more than three-quarters of brides were able to sign. There seems to be only impressionistic evidence before 1839: cf. E.P. Thompson, The Making of the English Working Class (Harmondsworth ed., 1968, cf. 1st ed., London, 1961), 782-8: "It is difficult to generalize as to the diffusion of literacy in the early years of the century."

[75] The Catholics were nearly 0.4 million. For discussions of the religious census of 1851, O. Chadwick, The Victorian Church, i (3rd ed., London, 1971, cf. 1st ed., 1966), 363-9; K.T. Hoppen, The Mid-Victorian Generation 1846-1886 (Oxford, 1998), 430-3; Harrison, The Early Victorians 1832-1851, 150-9.

[76] L. A. Tollemache, ed. A. Briggs, Gladstone's Boswell... (Brighton, 1984), 92, 202.

[77] Perhaps I build too much here on B. Askwith, The Lytteltons... (London, 1975), 113n, 130.

[78] Punch, 29 April 1908, 316. Poor Dolly "had not one domestic habit, / She could not mend a sock, or bake a tart. / Or even skin an ordinary rabbit – / Her parents simply spoiled her from the start." However, Dolly married into the aristocracy.

[79] Punch, 5 March 1913, 182; 26 June 1914, 482.

[80] Punch, 4 January 1911, 7. The verse lamely concludes "List my lay", which presumably meant "Hear my song". Its arithmetic was arbitrary, combining 1 + 9 +11 to make 3 x 7 (21). Naturally, the year was expected to run its course. "Nineteen-hundred and eleven! / Thus, when worn and wan with snow, / Multiple of three and seven, / Rhyming perfectly with heaven, / Out you go." By December, Britain would bid farewell to 1911 without regret.  A quarter of a century later, Punch rhymed "heaven" with "1937", but with less optimism. 

[81] "Is the sun yet cast out of heaven? / Is the song yet cast out of man? / Life that had song for its leaven / To quicken the blood that ran / Through the veins of the songless years / More bitter and cold than tears.…" Swinburne, The Last Oracle (1878).

[82]  In 1931, Sitwell evoked memories of childhood: "… with my doll and me in Heaven / Hear the nursery clock strike seven."

[83] In the final verse, the grieving mother is consoled by the thought that her son had died for Christ, and the two sacrifices become implicitly fused into one: "The Star stands forth in Heaven. / The watchers watch in vain / For Sign of the Promise given / Of peace on Earth again – / (Again! Again!) / 'But I know for Whom he fell' – / The steadfast mother smiled, / 'Is it well with the child – is it well? / It is well – it is well with the child!'"

[84] The Town Clerk, 1948.

[85] The Times, 7 June 1878. I have not traced the source.

[86] Lady Dorothy Nevill, an observer of Society manners, noted that of a letter from Disraeli in 1873 that he wrote "'a hotel' not 'an hotel'", and presumably used the same form in conversation: this she approved. However, aspiration incidentally involved a disruption of internal emphasis, to the dismay of purists. "Stress on the second syllable is correct," insisted the New Zealand expert on English, Professor Arnold Wall. "I hear 'HOtel', stressed on the first, with pain and amazement." As late as 1996, R.W. Burchfield noted that "a diminishing number of people in Britain pronounce hotel and humour with the h silent." R. Nevill, ed., The Reminiscences of Lady Dorothy Nevill (London, 1906),  208; (Christchurch) Press, 31 December 1954;  Burchfield, ed., Fowler's Modern English Usage, 631.

[87] A. J. Ellis, On Early English Pronunciation … (London, 1889), 227-30. This was the 5th volume of a study by Ellis linking Chaucer and Shakespeare to 19th-century usage, published for the Philological Society.

[88] Walker, Critical Pronouncing Dictionary, xiii.

[89] I should note here that, strictly speaking, the adjective should be "disyllabic", since the ultimate derivation of the term "syllable" is from Greek.

[90] Attempts made at the Last Night of the Proms in the 2020s to sing "Heaven" as two full syllables jar with the music.

[91] Walker, Critical Pronouncing Dictionary, 14, 297. He added "that though evil and devil suppress the e, as if written ev'l  and dev'l, yet that cavil and pencil preserve the sound of e distinctly; and that Latin ought never to be pronounced, as it is generally at schools, as if written Lat'n." It will be noted that here he regarded –i– spellings as conveying –e– sounds. 

[92] Smart, Walker Remodelled, xli. Quoting this in 1889, A.J. Ellis commented that "devil" is "is scarcely heard now but in the pulpit". It is not clear whether he referred to the alleged mispronunciation or to the general retreat of belief in Hell. Ellis, On Early English Pronunciation, 227.

[93] A Dictionary of the English Language (abridged Webster, ed. W. Wheeler, Springfield, Mass., 1872), xii.

[94] Otago Daily Times, 11, 14 November 1902, via the National Library of New Zealand's PapersPast online newspaper archive. For the Dictionary of New Zealand Biography on Burton: https://teara.govt.nz/en/biographies/2b51/burton-alfred-henry.

[95] "Ten Territorials fancied India fine; Till one caught malaria, and then there were nine. / Nine Territorials mourned his hapless fate; One found a cobra, and then there were eight. / Eight Territorials hoped he'd rest in heaven; One took his topee off, and then there were seven. / Seven Territorials brooded on their fix; One picked up dysentery, and then there were six." And so on. The topee, better known as a pith helmet, was a lightweight head covering that was believed to be necessary to protect Europeans from sunstroke. Punch, 15 September 1915, 226.

[96] Punch, 26 February 1936, 226.

[97] Punch, 23 December 1936, 707.

[98] As it was published by Oxford University Press, Henry Martin's book needs to be taken seriously. It seems to have been primarily intended to support the use of the English language in India. Martin's statement was curiously worded: "When words rime in two syllables, they are called double, or 'feminine' rimes ; e.g. sorrow, morrow; feather, weather; heaven, seven.)" His unconventional (and unexplained) spelling of 'rhyme' must have puzzled Indian students. H. Martin, The Teaching Of English Pronunciation (Oxford 1938), 123.

[99] Winterton's title was an Irish peerage, which made it possible for him to sit in the House of Commons.

[100] M. Foot, Aneurin Bevan: a Biography, i, 1897-1945 (London, 1966 ed., cf. 1st ed. 1962), 334.

[101] The poem was appropriately called Inland Waterway.

[102] Punch, ii, 1842, 255. St Mary Axe takes its name from a church that was demolished in the 16th century: the connection with an axe is obscure. Samuel Pepys once lived there. "Simmary" makes some sense as a slurred form of "St Mary". More curious was the Oxford nickname for St Mary Hall, "Skimmery", perhaps because it was a poor foundation. The nickname was once "a well-known substitute" but seems to have been forgotten after the absorption of St Mary Hall as part of Oriel College in 1902. Ward, William George Ward and the Oxford Movement, 24.