Why does beggar end in -ar? Many identically sounding words just use -er, if not all.


  1. bumper
  2. pepper
  3. tagger
  4. chanter
  5. pegger

They all use the -er version. Also, history shows that beggar was once spelled with an -er: begger.

Why did beggar adopt -ar over -er, and why haven’t other words done the same thing?

  • 2
    Google ngrams shows that the spelling with an a was pretty much always more prevalent. books.google.com/ngrams/… Commented Oct 5, 2014 at 13:09
  • 3
    It is pointless that the world evolved to a change from "begger" to "beggar" when the original was "begart". What is the advantage and how was this "ruled"? Commented Oct 5, 2014 at 13:22
  • 2
    Surely beggar is closer to begart? FWIW I don't pronounce pepper as rhyming with beggar. One is "er" as in ermintrude and the other "ar" as in "aromatic". Commented Oct 5, 2014 at 13:26
  • 3
    English evolved though. It wasn't designed by some committee. Commented Oct 5, 2014 at 13:36
  • 8
    @MartinSmith Are you really sure that you pronounce begger and pepper with a contrastively different final syllable? If so, you’re probably the first person I’ve ever come across that does so. The vowels in those two final syllables are slightly different, but that’s because they adapt to their preceding consonants: the /ə/ is retracted a bit after velar consonants, and fronted and raised a bit after bilabials. At the very least, tagger, pegger, dogger, logger, bigger, and begger should all have the exact same final syllable; so too bumper, paper, robber, and pepper. Commented Oct 5, 2014 at 16:37

4 Answers 4


Is it -er, -ar, or -or?

The first thing to understand about historical variation in modern beggar versus older begger is that the history of writing ‑ar, ‑er, ‑or for English word-endings is rather complex and not a little muddled. It is unwise to look for perfect predictability here.

In answer to your question about why other words didn’t do that, it turns out that some did. Like beggar, liar too started off with ‑er, as did pedlar, and even today both peddler and pedlar continue to occur even though the ‑er form is the elder. Going the other way, Scots once had socerar for sorcerer, and you can still find people who write sorceror.

It should be no great surprise that there would have been broad historical variation in spelling these sorts of words, considering how all three of ‑er, ‑ar, ‑or as unstressed final syllables are and were pronounced identically by most speakers. Before spelling was regularized, whether a written word ended in -er, ‑ar, or ‑or was up to each individual writer’s preferences, and those in the north of the Isle of Britain often used ‑ar here where ‑er was more prevalent in the south. A lot of the northern ‑ar words later got reworked into ‑er words, but not all of them. And some, like begger, went the other way.

How that sorted itself out under regularized spelling, let alone readjusted spelling by those trying to toe a more Latinate line, did not always follow the same path for each word. The French also had some hand in this, since today’s friar is spelled that way because despite having been a frater in Latin, we got the word from frere in Old French (Modern French frère).

Although the normal agent-noun suffix in English is ‑er, English also has a number of agent-nouns that derive from Latin now have ‑ar there instead, such as bursar, scholar, vicar; vulgar and even cellar are also from Latin. But those only settled out that way due to spelling reform; most were originally ‑er words in English because of having come to us through the French, who had changed Latin -arius words into -(i)er words. Some of those instead ended up looking like solitaire in English.

Thanks in part to the invention of Old French which had only ‑(i)er there, many of these words were once spelled with ‑er before getting put “back into” ‑ar form under 17th-century spelling reforms. And some — like pedlar, liar, and beggar — seem to have been dragged along for the ride more by analogy rather by etymology.

While these ‑ar words that we got from Latin (with or without French intervention) are in theory distinct from native English words ending in ‑er and from Latin ‑or words (mostly agent-nouns like author, cantor, doctor, censor, cursor, elector, inventor, lictor), this distinction was not always well-preserved: notice how both sorcerer and sorceror occur, as do both imposter and impostor, with sorcerer and impostor now the more accepted or common renditions of those pairs. Plus even though ‑er was usually a native-English ending, Latin also contributed some ‑er words of its own to English, like neuter, integer, dexter, sinister, super.

Because of how 17th century orthographers wanted to make words look more like Latin when writing them, eventually some of our words that were normally ‑er even up north got reworked into ‑ar words instead, consciously or unconsciously tying them to ‑ar Latin forms whether real or imagined.

Examples already mentioned include liar and pedlar, but there are many more. A lot of words had their standardized spellings changed into unhistorical forms during this time, famously including ones like island and debt. In its article on English Spelling Reform, Wikipedia states:

From the 16th century onward, English writers who were scholars of Greek and Latin literature tried to link English words to their Graeco-Latin counterparts. They did this by adding silent letters to make the real or imagined links more obvious. Thus det became debt (to link it to Latin debitum), dout became doubt (to link it to Latin dubitare), sissors became scissors and sithe became scythe (as they were wrongly thought to come from Latin scindere), iland became island (as it was wrongly thought to come from Latin insula), ake became ache (as it was wrongly thought to come from Greek akhos), and so forth.

As the last sentence cited above shows, mistakes were made during this time. And while it is arguable, beggar seems to have been one of these. The OED is not completely certain of this, as they say “probably imitating”. In particular, they say of beggar:

The spelling in -ar has been occasional from 14th c., but the usual form in 15–17th c., as an ordinary agent-noun from beg, was begger: see ‑ar3.

Which states:

‑ar, suff.3, casual variant of ‑er, ‑or, suffix of agent, and ‑er suffix of comparative. Very common in north. dial., as syngar singer, forebear predecessor, soutar sutor; hear higher. And in modern Eng. in beggar, liar, pedlar. Probably imitating the refashioned scholar, vicar, pillar for earlier scoler, viker, piler: see ‑ar1, and ‑ar2, above.

The ‑ar1 case contains such words as altar, collar, pillar, solar, lunar, regular, similar, and so includes words that came to us both directly from Latin and via Norman French, and in English sometimes showed up as ‑(i)aire as in ordinaire and which are related to the ‑ar2 case.

The ‑ar2 case is words we refashioned from Old French ‑ier, but which ultimately have the same origin as ‑ar1 words. These include words like bursar, mortar, vicar. Many of these used to be ‑er words in English, but got redone in a “more Latin way” to turn them into their current ‑ar forms.

The histories behind ‑er and ‑or words in English are both of them even more complex than those of ‑ar are.

  • Note that the -ar forms are derived from adjectives and are not primarily agent nouns: Latin never had -ar as such: the ending is a blend of ① the highly productive adjective suffix -ārius (-ārium), and ② the less prevalent, but still quite common, adjective suffix -āris (-ārem), a dissimilated form of -ālis (-ālem) for roots containing an l. Both just vaguely indicate ‘of the kind specified; relating to’, whereas the -or nouns are straight from Latin and are truly agentive, even in Latin itself (and in PIE, where the main agentive suffix was -tor). Commented Oct 5, 2014 at 16:49
  • @JanusBahsJacquet Yes, I’ve tried to fix that, since I was misstating that entire agent-noun issue from the Latin perspective. Cursors and actors are different from something that was similar or, thanks to French, solitaire and extraordinaire.
    – tchrist
    Commented Oct 5, 2014 at 17:22
  • Still, if the word comes originally from Middle Dutch beggaert, that implies that it doesn't originate with the English verb to beg, but that the verb was derived from the noun.
    – Robusto
    Commented Oct 5, 2014 at 18:04
  • I would have stopped at the first paragraph if I were writing this. But nonetheless, this gets an upvote from me.
    – Pitarou
    Commented Oct 6, 2014 at 11:46
  • @Pitarou I think the problem is that the OP hasn’t expressed why he thinks beggar should be spelt as begger. It probably was phonetic when we first got it, but as it got used more and more its final syllable fell to /ɚ/, so people started writing it that way. It’s like how calendar and solar now have /ɚ/ due to long use in the language, but exemplar and seminar do not (well, plus they’re stressed and therefore unreduced, but it’s all related).
    – tchrist
    Commented Oct 6, 2014 at 12:31

Well, the vagaries of English spelling are legendary. This is one of its more peculiar ones.

According to Etymonline.com,

c.1200, from Old French begart, originally a member of the Beghards, lay brothers of mendicants in the Low Countries, from Middle Dutch beggaert "mendicant," of uncertain origin, with pejorative suffix (see -ard). Compare Beguine. Early folk etymology connected the English word with bag. Form with -ar attested from 14c., but begger was more usual 15c.-17c.

So it started off with the -ar, got turned around to -er, and went back again. There is no good reason for it, it just turned out that way.

  • 6
    Who says it has to make sense? If there is one thing we can be sure of, it's that English spelling has very little "sense" to it.
    – Robusto
    Commented Oct 5, 2014 at 13:21
  • 5
    There was a movement during the 17th century to institute "etymologically correct" spellings. This is when the 'b' was restored to debt (whose origin was Latin debitum, but it had been pronounced without a 'b' since before English borrowed the word from Old French in the 14th century). I expect the 'a' was restored to beggar for the same reason. Commented Oct 5, 2014 at 13:47
  • 1
    I might choose 'begger of questions' over 'beggar of questions', but I'd probably rephrase. Choosers can have 'beggers'. Commented Oct 5, 2014 at 14:59
  • 1
    @bye What “uncouth neologism” do you purport for the verb beg? Middle English had “Scheome ich telle uorte··beggen ase on harlot··his liueneð.” in the early 1200s, and Old English in the late 800s had “Hit is swiðe wel be ðæm ʒecweden ðæt he eft bedeciʒe on sumera, & him mon ðonne noht ne sell.”
    – tchrist
    Commented Oct 5, 2014 at 18:22
  • 2
    @tchrist Agreed completely with your point (though not necessarily with your last quote—it’s not really certain whether beg is the continuation of bedecian or not, after all). If we start referring to words that appeared almost a millennium ago as “uncouth neologisms”, we’ll have to come up with a whole spectrum of new words to describe words that can actually still be considered ‘new’ (such as neologism, for example, which is somewhere around 600 years ‘newer’ than beg as a verb). Commented Oct 5, 2014 at 18:56

Willard Van Orman Quine, the great polymath, on this topic (Quiddities, p115-116):

Examples [of historical relics in English that can reward contemplation] are evident at every turn. Let me cite one of the less evident ones: the ending -ar in beggar, burglar and pedlar. In our words of Old English origin the usual ending for agent is -er, and in words borrowed from Latin it is -or; what then of this deviant -ar? Looking up origins, we find that historically beggar, burglar and pedlar are not even formed from verbs; there was no question of an agent ending. The three nouns came first, from other sources. Afterward we derived the verbs beg, burgle and peddle from them by "back-formation", subtracting the -ar as if it had been an agent ending.

  • Middle English had a verb beg in the early 1200s: “Scheome ich telle uorte··beggen ase on harlot··his liueneð.” If beg is a back-formation, it is an extremely old one. One can find the verb beguigner in French starting from the 1300s. This is likely why the OED says of to beg that “Perhaps the most likely derivation is from the OF. begart, begard, and begar, med.L. begardus = Beghard, or its synonym beguin, beguin, and deriv. vb. beguigner, beguiner ‘to act the beguin.’”
    – tchrist
    Commented Oct 5, 2014 at 18:28

@Robusto has the answer, however, to magnify a little, beggar is probably not, as we might assume, derived from the verb "to beg" as with many of the other examples you give. On the contrary, the verb is derived from the noun. So normal formation would not be expected. Also from Etymonline:

beg c.1200, perhaps from Old English bedecian "to beg," from Proto-Germanic *beth-; or possibly from Anglo-French begger, from Old French begart (see beggar). The Old English word for "beg" was wædlian, from wædl "poverty."...

A parallel might be helpful. We have the word "burglar" as someone who breaks in an steals. In the UK the verb that is used for this action is "to burgle" -- as in "my house was burgled." However, in the USA they generally use "to burglarize", as in "my house was burglarized."

The difference derives from the fact the the original word is the noun, and the verbs are back formed from it using two different methods of back formation.

The history of the word is pretty interesting.

burglar 1540s, shortened from Anglo-Latin burglator (late 13c.), earlier burgator, from Medieval Latin burgator "burglar," from burgare "to break open, commit burglary," from Latin burgus "fortress, castle," a Germanic loan-word akin to borough. The intrusive -l- is perhaps from influence of Latin latro "thief" (see larceny). The native word, Old English burgh-breche, might have influenced the word.

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.