Looking at a medium-sized word list, all words are written "erence" with the exception of "deterrence" and the name "Terrence". I can find 50 words ending in "erence".

What is different about deterrence? Does it have an odd etymology? Or is it used enough to maintain the odd spelling?


Here are some words that end in rence taken from my list:

adherence, belligerence, circumference, coherence, coinherence, conference, countertransference, decoherence, deference, difference, imperence, incoherence, indifference, inference, inherence, interference, irreverence, misinference, misreference, nonadherence, nondeference, noninterference, nonpreference, nonreference, pre-adherence, preconference, preference, pre-inference, pre-interference, prepreference, pre-reference, proconference, reference, reinterference, retransference, retrotransference, reverence, semicircumference, semidifference, subconference, subreference, superindifference, superinference, surreverence, Terence, therence, transference, unadherence, unindifference, unreverence.

  • 1
    Have you looked at Etymonline or another etymological dictionary? It seems to come from the Latin de+terrere; compare "terrible". Can you indicate what similar words you're considering, and what you don't understand about their etymology?
    – Stuart F
    Apr 29, 2022 at 9:34
  • 1
    No I hadn't looked into the etymology of deterrence. I asked the question here instead. The question is really why is deterrence different from every other word ending in "ence", so it is difficult to understand all of their etymologies, but I agree looking at some examples would be helpful.
    – Att Righ
    Apr 29, 2022 at 9:42
  • 1
    Terence is usually spelled with one 'r' (but people are free to spell their children's names any way they choose). Apr 29, 2022 at 10:02
  • 4
    I don't think appeals to etymology are particularly helpful here. Sure - deter has its roots in Latin deterre. But the first "similar" word I just checked was defer, which comes from Latin differre, which also has the double /r/. It may not be the whole answer, but I think it's relevant that deterrence has stress on the syllable immediately preceding the doubled letter, whereas most similar words normally just have a neutral schwa there. Apr 29, 2022 at 10:50
  • 2
    ...I also think it's relevant that if I needed a term for "the act of deferring" (as in delaying, rather than respectfully ceding authority) and for some reason I didn't want to use the word deferral (double /r/, obviously), I'd prefer to spell that one deferrence. That would reflect the fact that I'm pronouncing it with stress on the second syllable for that contrived meaning (whereas stress goes on the first syllable for the standard "respectful" sense). Apr 29, 2022 at 11:00

2 Answers 2


The spelling of deterrence is largely about its Latin etymology. The same goes for most (although not all) -ence words in general, as well as -ent, -ant, and -ance words.

The position of stress could have also have some relevance, but the stress difference between deterrence and inference etc. is itself most likely based on the different etymologies of the words. (In words fully formed within English, such as all words ending in -ing e.g. deterring, inferring, preferring, the position of stress in the base English verb is certainly relevant.)

deterrence comes from a Latin verb with the stem deterre-

The Latin verb that deterrence comes from is deterrere, first person singular deterreo, present participle deterrens. The stem has a double -rr- in it, so all forms of the verb have -rr-. We also find -rr- in concurrence, recurrence, occurrence from the Latin verb currere, curro.

The Latin verb that deference comes from is differre, first person singular differo, present participle differens. Although the infinitive form differre has a double rr, this double consonant is not actually part of the verb's stem: rather, the stem ends in only one r, and the second r comes from the infinitive suffix -re being added directly after the stem. That is, the infinitive is built as difer- + -re just as the infinitive deterrere is formed as deterre- + -re. Therefore, forms of differre other than the infinitive, such as the first person singular and the present participle differens, only have one r.

So the spellings of these words with rr and r respectively are both regular from the perspective of their etymologies.

The etymological sources of the list of -erence words

As you may have noticed, most of the words on your list can be grouped according to a small set of endings. The ones ending in -ference mostly come from ferre, fero, which was a common verb in Latin with many prefixed versions in use. (Interference is an exception: it's from a different source, and it is not built according to Latin rules of word formation.) The ones ending in -herence come from haerere, haereo. The point I'm getting at is that from the perspective of etymology, these are not all independent data points. If one of those two Latin verbs had happened to have a stem containing -err-, you'd see a lot more -errence spellings in English.

  • Spot on. What this illustrates is the statistical fallacy: the statistic seems orthographically significant, until you take into account the morphology. What is more, 'fero', along with 'sum', 'volo' and 'malo' are defective. 'Sum' is crazy, like our own copula (perhaps reflecting earlier origins). They all have a doubled r/l/s in the present infinitive (ferre, velle, malle, esse); but the participles - apart from sum - are like all 2nd or 3rd declension verbs (ferens, volens, malens) with a single 'r'/'l') The noun from 'ferens' changes 's' into 't' and adds - 'ia'. [continued]
    – Tuffy
    Apr 30, 2022 at 10:55
  • [continued] This makes it 'ferentia', which has never, as far as I know, been used EXCEPT in all its prepositional compounds, of which there are great hosts, because the verb 'fero' is so flexible (vague) in meaning. As to the others, these follow the morphological rules: 'Reverence' from 'vereor' (I fear), 2nd declension: participle 'reverens', abstract noun changes 's' to 't' and adds 'ia' giving 'reverentia'. Most words in the list in the question are compounds of the same basic word.
    – Tuffy
    Apr 30, 2022 at 12:40

I can think of only three three-syllable words that have the main accent on the second syllable and end in (to use Merriam-Webster's pronunciation symbols) the sounds "ərən(t)s": deterrence (which MW dates to 1861), occurrence (1539), and transference (1681). So from one point of view, the odd spelling out isn't the double-r deterrence but the single-r transference.

The earliest so-called pronouncing dictionary that I am aware of, William Johnston, A Pronouncing and Spelling Dictionary (1764) has an entry for occurrence (given as occúrrence), but not for deterrence (as one would expect) or transference (as one might not). The closest related terms that Johnston gives to those two terms are detér and detérred in the vicinity of deterrence and tránsfer, transfér, and tránsferable in the vicinity of transference. Clearly Johnston isn't advising readers about how to pronounce the words at the consonant-and-vowel level but only which syllable to emphasize.

Although Johnston doesn't explain the two pronunciations given for transfer, most English speakers today (and perhaps in his day) use tránsfer for the noun and transfér for the verb. Strikingly, however, most English speakers today—at least in the United States—accent the second syllable of transferable, not the first; in fact, Merriam-Webster's Eleventh Collegiate Dictionary (2003) gives only the pronunciation "tran(t)s-'fər-ə-bəl" for the word, although it lists transferrable as a variant spelling of transferable. Johnston's treatment of transferable suggests that at some point in its early existence transference may have been pronounced with the accent on the first syllable, which would then make its single-r spelling consistent with such other three-syllable English words as conference, deference, difference, inference, preference, reference, reverence, severance, and temperance.

Likewise, James Barclay, A Complete and Universal English Dictionary on a New Plan (1774) provides a pronunciation (or accented syllable) for occurrence ("occu'rrence") but nothing for deterrence or transference.

John Walker, A Critical Pronouncing Dictionary, and Expositor of the English Language, third edition (1802) offers the following entry for transferable:

TRANSFERABLE trans-fer'-a-bl, or trans'-fer-a-bl. a. Capable of being transferred. | I have met with this very common and useful word in no Dictionary but Entick's, where the accent is very properly placed on the second syllable ; as all words of this form ought as much as possible to retain the accent of the verb from which they are derived.

The second edition of Walker's dictionary (1797) has no entry for transferable, although it does indicate that "trans-fer'" is the emphasis proper emphasis for the verb form of transfer and "trans'-fer" is the proper emphasis for the noun form. In the sixth edition of Walker's dictionary (1809), however, the entry lists both transferable and transferrable as acceptable spellings, each with its own pronunciation:

TRANSFERABLE trans'-fer-a-bl, TRANSFERRABLE trans-fer'-a-bl, a. Capable of being transferred.

The entry then repeats the 1802 edition's comment about the propriety of retaining the accent used for source verbs in derived adjectives "of this form."

Noah Webster, A Compendious Dictionary of the English Language (1806) spells transferrable with a double r and puts the accent on the second syllable. However, Webster's massive A Dictionary of the English Language (1828) changes the spelling to transferable while retaining the accent on the second syllable. The 1840 edition of Webster's dictionary sticks with transferable, but it also includes for the first time an entry for transferrence (spelled with a double r:

TRANS-FER'RENCE, n. Act of transferring.

The next major edition of An American Dictionary of the English Language (1864) includes two spellings of the word, with a different accent for each:

Trans'ference, n. Act of transferring; transfer.


Trans-fer'rence, n. See TRANSFERENCE.

Evidently there was still some disagreement about how the word should be spelled and pronounced, with Webster's now favoring the transference/trans'ference option. The 1864 Webster's dictionary is also notable for marking the debut of deterrence:

De-ter'rence, n. That which deters; a deterrent; a hinderance. {Rare.}

Just four years earlier, Joseph Worcester, A Dictionary of the English Language (1860) had listed transferrence as the only spelling for that noun but had omitted any mention of deterrence (although it did include an entry for deterrent).

Webster's International Dictionary (1890) retains the two spellings and the two pronunciations of transference/transferrence without change; but Webster's New International Dictionary (1909) lists only the spelling transference, giving it a variable pronunciation:

transference (trăns'fẽr-ĕns; trăns-fûr'ĕns), n. Act of transferring; conveyance; passage; transfer.

This remains the order of presentation (and pronunciation preference) through the fourth edition of the Webster's Collegiate Dictionary series (1931). The Fifth Collegiate 1936), however, flips the pronunciation preference:

transference (trăns-fûr'ĕns; trăns'fẽr-) n. 1. Act of transferring; conveyance; passage; transfer; as thought transference (i.e., telepathy). 2. Psychoanalysis. Direction of feeling and desires, esp. as unconsciously retained from childhood, toward a new object.

That order of pronunciation preference has remained unchanged in editions six (1949) through eleven (2003) of the Collegiate Dictionary series.


The spelling and the accented second syllable of deterrence have been treated consistently in U.S. dictionaries since the word first appeared in the 1864 edition of Merriam-Webster's An American Dictionary of the English Language.

The history of transference is quite different. It first appears in a U.S. dictionary in the 1840 edition of Webster's unabridged dictionary with the spelling transferrence and an accented second syllable (however, this same dictionary spells transferable with a single r, despite accenting the second syllable of it as well). Then, from 1864 to 1908, Merriam-Webster's dictionaries list two forms of the word: transference, with the first syllable accented, and transferrence, with the second syllable accented.

In 1909, Merriam-Webster drops the entry for transferrence and moves the accented-second-syllable pronunciation to the entry for transference, while keeping the accented-first-syllable pronunciation in the primary position. Finally, in 1936, as use of transference in psychoanalysis receives attention in its definition for the first time, MW flips the order of preference to put the accented-second-syllable pronunciation first.

This record suggests that the inconsistent combination of spelling and pronunciation for transference/transfer'ence (versus deterrence/deter'rence and occurrence/occur'rence) may reflect, at least in part, the longtime alternative pronunciation preference for trans'ference during the critical period when the spelling was still being hashed out. As for the double r in deterrence, that spelling is consistent with the the accented second syllable of the word, which dictionaries have consistently reported since they first began including entries for the word.

Side note: Instances of 'deterrence' from the 1700s

Although Merriam-Webster give a first occurrence date of 1861 for deterrence, the word began appearing in print long before that year. A Google Books search turns up three instances from the eighteenth century.

From Thomas Cooper, "Sketch of the Controversy of the Subject of Materialism" (read at the Manchester Society, January 17, 1787), reprinted in Tracts, Ethical, Theological and Political (1789):

  1. The question has been discussed almost 2000 years : during which time, every thing that human ingenuity could suggest in defence of the separate existence of the soul, has been advanced. The opinion moreover had been supported, not only by the pecuniary emoluments, and the honours of establishments, but by the arm of the secular power. Every motive that men could have to adopt it, has been applied, and every motive of deterrence has been exhibited. Yet have some of the wisest and the best men of the present day among us, and not only so, but members of the established church, and of the highest rank therein, remained unconvinced, that this dogma of immaterialism is so uncontrovertibly evident as inn general has been supposed.

From "Labour and Solitude More Rational Punishments Than Executions," in in The Gentleman's Magazine (February 1788):

The great defect in the punishments of this, and perhaps every other country, I apprehend to be this, that they answer very imperfectly either of the purposes for which punishments are principally intended, the reformation of the offender, or deterrence from the offence. The noise and bustle of a public execution, and the crimes which are committed at the very foot of the gallows, are too evident proofs, that little impression is made upon those for whose benefit it is particularly appointed.

And from "The Trial at Large on an Action for Damages Brought ... by the Righ Hon. George Fred. Earl of Westmeath, Against the Honorable Augustus Cavendish Bradshaw, for Adultery with the Right Hon. Mary Anne, Countess of Westmeath" (February 20, 1796):

It would appear in evidence, that when she [the Countess] had been driven from her own door in this carriage under a pretence of going to visit her friends, that instead of going to those friends, she had ordered her coachman to drive with the blinds up, to a remote part of the country, with Mr. Bradshaw, whom she took into her carriage on the way. These facts with a number of others would come out in evidence, and to that evidence he [the Solicitor General] would rather refer the gentlemen of the jury than by anticipating the detail—seem to aggravate a case which in itself was but too strong, and which he was confident must have that weight with respectable men bound upon their oaths to give an impartial verdict, and such an one, as while it made some reparation to the wounded feelings of his noble client, (for no sum of damages however great would make adequate atonement) would at the same time hold out an example for the deterrence of similar crimes and make some satisfaction to the violated laws of society.

  • 1
    Another pronunciation detail with regard to “deterrence” is that the vowel in the stressed syllable can be either the “merry” vowel, as in errant, aberrant, and terror, or the “furry” vowel, as in deterring.
    – herisson
    May 9, 2022 at 19:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.