The origin of nothing
Why negative words start with ‘n’
Most words don’t seem to have an inherent connection to the things they refer to.
For instance, there’s nothing in the sequence of sounds we use to write the word chair that suggests that estimable four-legged piece of furniture, beyond the convention among English speakers that it is called, of all things, chair.
Chairs could be called something else, anything else. A chair by any other name would be a seat.
The arbitrariness of the pairing between word and meaning is one of the key characteristics of human language. But language is not always so arbitrary. There are some words whose sounds do show a hint of inherent meaning.
Think, for example, of all the words in English beginning with gl- which have something to do with light: glow, glare, gleam, glisten, glitter.
Or the association of words beginning in sn- with the nose or mouth: snot, snort, snore, snarl, sniffle, snuffle, sneeze, and so on.
There’s also a connection between sl- and smooth motion, such as slip, slide, slick, sled, etc., as well as negative qualities, as in slum, slur, slog, slump, slouch, slovenly, and so on.
These not-entirely-arbitrary associations between meanings and sound sequences are called phonaesthemes. They keep nonsense poetry like Lewis Carroll’s Jabberwocky hovering just on the edge of comprehensibility:
‘Twas brillig, and the slithy toves
Did gyre and gimble in the wabe:
I don’t know what toves are, but the fact that they can be slithy implies that they likely move smoothly and may be somewhat sinister.
Armed with this knowledge, we might glance at the English lexicon and posit yet another phonaestheme in the language: an association between n- and negation.
Beyond the word negation itself, we have no, not, nor, none, never, neither… the list goes on.
But this is not the case.
The reason so many negative words begin with n- is because they all originate from one single word in the distant prehistory of the English language, along with the special propensity of this word to attach itself to other words.
But we don’t have to go back into the mists of time to understand the processes at play. In fact, there are clues all around us, if you only know where to look.
You’re reading The Dead Language Society. I’m Colin Gorrie, linguist, ancient language teacher, and your guide through the history of the English language and its relatives.
Subscribe for a free issue every Wednesday, or upgrade to support my mission of bringing historical linguistics out of the ivory tower and receive two extra Saturday deep-dives per month.
If you upgrade, you’ll be able to join our ongoing Beowulf Book Club. You can also catch up by watching our discussion of the first 1962 lines. Our next meeting is Thursday, December 4.
Nobody, nothing, no one
Our first clue can be found in the existence of words like no one, nothing, and nobody, which wear their history on their sleeve: they began life as combinations, or compounds, of two separate words no + one, no + thing, and no + body.
The spelling and pronunciation of no one still shows this origin very clearly: you might even consider no one two separate words, although dictionaries generally give it its own separate entry.
For nothing and nobody, on the other hand, the two parts are usually written together. This is probably because the pronunciation of the parts has changed: the no- in nothing no longer sounds like the independent word no, and the -body in nobody no longer sounds like body (for many speakers, anyway) when it’s said alone.
Just as older mountains have softer, more rounded features than younger mountains due to the effects of erosion over time, so too are the effects of phonetic erosion most clearly seen in older compounds.
Compare school board (first attested in 1833), in which school and board both sound like they do when said separately, with cupboard (first attested around 1380), in which neither element cup- nor -board is pronounced the way you’d say the individual words.
In fact, words like cupboard (or starboard, from steer board) have been so far eroded that, for most speakers, they are no longer identifiable as compounds.
But there was a time when each of these compounds was new. The oldest of the no one, nothing, nobody trio is nothing: it first appears in Old English, that is, before the year 1100.
At its birth, nothing, or, better, no thing, was two separate words. In fact, each of the two parts, no and thing — or, in Old English, nān and þing, took separate endings, such as the ending -es, which appears on both words in the following example:
On þām dæġe ġē ne biddað me nānes þinges.
‘On that day you will not ask me for nothing.’ (West Saxon Gospels, John 16: 23)
You may notice from the translation of this sentence that Old English used “double negatives”, such as not ask me for nothing. This is worth an article of its own, so I won’t discuss it any more today beyond saying this: yes, these double negative forms, though today considered nonstandard, were in fact present in Old English. And not only were they present; they were common in even the most formal language.
For many centuries after the Old English period, the words that would become nothing occasionally started to be written together as one word, but one-word and two-word forms coexisted until the 16th century (into English’s Early Modern Period), when the single-word form nothing eventually won out.
At some point during this process, the pronunciation of the no in nothing diverged from its pronunciation when said alone. We know this had occurred by the 18th century because we have pronunciation manuals which tell us that the modern pronunciation of nothing was standard by then.
Nobody came later than nothing, first appearing in the 14th century, when it too was most often written as two separate words, no body. By the 17th century, it began to be written often with a hyphen, as no-body. The present spelling nobody, without space or hyphen, became common in the 18th century.
The word no one was an even later addition to the English lexicon, first appearing in the 16th century. Judging from how it is written, with a space between the elements, no one still isn’t fully perceived as a single word (other than, perhaps, by linguists and writers of dictionaries). There are some writers today who spell no one with a hyphen, as in no-one, a practice which began in the 19th century.
As we saw with nobody, writing a compound word with a hyphen is often a step towards writing it as a single word, which suggests that no one is well on its way to single-word status in the minds of at least some English writers.
The words no one, nothing, nobody did not spring up in isolation. Other, similar words came along, such as someone, something, somebody and anyone, anybody, anything, which were formed out of compounding the same nouns one, thing, and body, this time with some and any.
Forming words by compounding is just something that English likes to do. And it’s been doing it for a long time, in particular with the negative. In fact, there are many more negative compounds lurking in the English language than the ones we can easily identify, including a word as basic as no.
Why we say “none other than”
No is one of the first English words acquired by adults and children alike — even if adults might sometimes wish their children learned it just a little bit later. It’s the element that English ended up using to create negative compounds such as the trio we discussed above (no one, nothing, nobody).
All this might tempt you to imagine that no is the basic, original negative word in the English language, but that would be too easy.
No is a compound too, just one built from the original negative word in English, the tiny little word ne.
This word ne was the first part of the compound: for the second part, it depends on what part of speech we’re talking about. The Modern English word no descends from two once-separate Old English words that later collided: one an interjection (as in no, I don’t think I will) and the other an adjective (as in no country for old men).
The interjection no — that is, the opposite of yes — derives from an Old English word spelled nā or nō. It was originally an adverb, used primarily to negate anything other than a verb,1 much in the same way that Modern English uses the word not in phrases like not now or not so long ago. It was also used as an interjection, like Modern English no (as in no, thanks) in Old English times.
The forms nā and nō merged into the single form no in southern dialects of Middle English, and from there it came into Modern English as the word we say when we don’t want something: no!
These words nā/nō has its origins as a compound of ne, the actual basic word meaning ‘not’ in Old English, and ā (also spelled ō) ‘always.’ The word ā doesn’t survive into Modern English, dropping out by around 1600.
Maybe it was just too insubstantial to survive, consisting as it did of only a single vowel. It has since been replaced by words like ever and always.2
At any rate, our Modern English interjection no, at its root, was an emphatic negation, meaning ‘not ever!’
The other word that gave birth to Modern English no is Old English nān, originally a compound of ne and ān ‘one.’ We saw this word nān above, in the phrase nānes þinges ‘of nothing.’
In the later Middle Ages, the ā vowel in this word nān became an o-vowel (as we learned when discussing what holy has to do with Halloween) and, for reasons that are still unclear, the resulting word none eventually acquired the strange pronunciation that it has today.
This word nān, and its descendant none, was used as either an adjective (alongside nouns such as þing ‘thing,’ mann ‘person,’ wundor ‘surprise,’ and so on) or as a pronoun, that is, on its own, as in the modern phrase and then there were none.
The word none, of course, survives into Modern English, but only as a pronoun. There is one exception: none can be used as an adjective in the phrase none other than…. The reason that it survives in this phrase and not elsewhere is because (a) it’s a fixed expression, which tends to stick around, and (b) because other starts with a vowel.
This use of none before vowels hearkens back to the late Middle Ages, where the word none was used before vowels (or h) and an n-less variant no was used before consonants.
Does this remind you of another English word? It makes me think of how a vs an works.
To this day, we say a banana but an apple, because an is used before words beginning in a vowel. And this is no accident, because a/an comes from the Old English word ān, that is, the non-negated form of no/none!3
After around the year 1600, this system broke down, for no vs none, anyway, and no started to be used before vowels as well, leaving us with the situation we have today, where no is used as the adjective and none as the pronoun.
So no and none are ancient compounds of ordinary words, like ā ‘always’ and ān ‘one,’ along with a more basic negative element ne.
And it’s this negator ne — which no longer exists outside of its many compounds — that was the ordinary way of negating a sentence in Old English. This is the form that links English negation with negation in the many relatives of English, including Latin, the very language that gave us the word negation. (Yes, it’s the same ne.)
It’s not just English…
If you think of how archaic English formed negatives, you may call to mind the form that often appears in Shakespeare’s Elizabethan English: I know not. This isn’t what we do in Modern English. We prefer to add a dummy verb do, and we negate that: I don’t know.
But Old English had a third strategy, one that predates either I know not or I don’t know. This strategy was simply to place ne before the verb. Often, the subject would then move after the verb, due to a quirk of Old English word order we’ll talk about another day.4
So the positive sentence:
Iċ ete berġan.
‘I eat berries’
corresponds to the negative:
Ne ete iċ berġan.
‘I don’t eat berries.’
This little negative word ne has an illustrious pedigree. It is a direct descendant of the negative word in Proto-Indo-European, the ultimate ancestor of English, Latin, Ancient Greek, and many other languages besides. This original negative word has been reconstructed as *ne, where the star indicates that it’s a reconstructed word, rather than one we find in a text.
This same negative word was passed into the descendants of Proto-Indo-European, including other distant relatives of English like Latin. In Latin, however, the basic negative word is nōn. It starts with an n-sound, but it has a different ending.
By now, however, we’re becoming used to this pattern: nōn is yet another ne compound! It’s most likely composed of ne joined to the ancestor of ūnus ‘one,’ making Latin nōn exactly equivalent, in a formal sense, to the English none, which also comes from a sequence meaning ‘not one.’ Small world!
Although *ne did not survive directly in Latin, it left behind traces in words such as nōn, neque ‘and not,’ nūllus ‘not any,’ nihil ‘nothing; lit. not a bit,’ nēmō ‘no one; lit. not a person,’ numquam ‘never; lit. not ever.’5
Even the Latin word negāre ‘to deny’ — the source of our English word negative — is simply ne turned into a verb, ‘to ne’!6
Returning to nōn, we see that some of its descendants had similar careers to its English analogue none. A good example comes from Spanish: the Latin nōn is the ancestor of the Spanish negator no, which has, over the years, lost its final -n.
The history of the Spanish negative shows another surprising parallel with the history of the English adjective none. Recall that none first lost its final n-sound in certain positions (before consonants), and only later generalized the n-less form. The result is the Modern English form no.
This is, in broad strokes, exactly what happened to the Spanish negator no: it began as non, which first acquired in the Middle Ages a variant n-less form no in certain positions,7 in particular before l. Then, after a convoluted history in which people went back and forth between no and non in writing, the n-less forms spread to the entire language, leaving only no. It’s not just Spanish, either: the French negator ne descends from Latin nōn, through a similar medieval pair nen/ne.8
Seeing these uncannily parallel developments in two separate language families, one starts to wonder: Why do these negative words have a habit of forming compounds with other words and then disappearing?
Really, you can’t eat those berries
There are probably two, interrelated reasons. One is that ne is very light: just a tiny little syllable, with a tiny little short vowel, that comes before a verb. It’d be easy to miss, especially in a noisy environment, kind of like how hard it is to hear the difference between can and can’t.
But the difference between negative and positive sentences is usually an important one: if you hear you can eat those berries! when the speaker meant you can’t eat those berries!, that could have some negative consequences for you.
Probably as a result of the high-stakes nature of negation, negative sentences tend to attract emphatic words and phrases that emphasize the negation: think of how phrases like at all and a bit combine with Modern English negation to give sentences like I can’t see at all or I didn’t eat a bit.
Old English was like this too (as was prehistoric Latin, but I’ll limit the discussion to English, because we have much more data): ne is such a light form of negation that it ended up needing some reinforcement with forms like ā ‘ever’ or ān ‘one.’ Out of this came the negative compounds we’ve been discussing today, such as nā ‘not ever’ and nān ‘not one’.
Speaking of not, it too is the result of one of these combinations: it comes from ne + āwiht ‘anything,’ itself a compound of our familiar ā ‘ever’ and wiht ‘thing; living creature.’
This word wiht, by the way, is the source of the word wight, which the folklorist Andrew Lang, and later, J. R. R. Tolkien, used to spooky effect in the phrase barrow-wights, referring spirits which dwelt in barrows, or burial mounds.
The result of all this compounding was the Old English word nāwiht ‘nothing.’ But even in Old English, it started to be worn down, appearing in forms like nāht or nōht:
Næfde hēo nōht on hire.
‘She (the soul) didn’t have nothing on her.’ (Blickling Homilies)
Over time, this shortened form nāht/nōht became the word naught or nought, meaning ‘nothing.’ When it got worn down further, it lost its gh and became our more familiar negative marker not.
As we see in the quotation above, these emphatic negative compounds originally occurred alongside the basic negator ne,9 just as we saw in the Gospel quotation earlier. The result of this is the “double negative” construction. But, strictly speaking, once you’ve got into the habit of emphasizing your negative sentences with these negative compounds, only one of the negatives is truly necessary. And it’s the older, lighter one that tends to drop out.10
But there was a long period, lasting the length of the later Middle Ages (1100–1500), where ne and other negatives coexisted. Eventually, however, ne dropped out of the language altogether. It was more or less completely gone by the year 1600, although it hung around later in poetry, certain set phrases, and when people were trying to sound particularly old-fashioned.
Today, only its compounds remain, and there are many of them. I didn’t even mention neither, nor or never, although, by now, you can probably figure out their origins easily enough!
So this is the origin of the association between n-words and negation in the English language. It’s not that n- is a phonaestheme evoking the idea of negation in the way gl- evokes the idea of light. Instead, these negative n-words all look alike because they are all descendants of one single word ne, a word too tiny to survive the ravages of time.
But ne left us something to remember it by: a collection of negative compounds, many of which have themselves been eroded by time to the point that we can no longer recognize them as such.
One day, no doubt, our word not will itself come to naught. In fact, nothing could be more certain: not is already reduced to -n’t in many words. How much longer will not hold on before it drops out altogether, with some formerly emphatic word like a bit or at all coming to replace it?
Time comes for all words, if you say them often enough.
Technically, nā/nō negates anything other than a finite verb. In fancy linguist terms, it was used for constituent negation rather than sentential negation.
Ā survives, hidden in its replacement ever, which was itself originally a compound ā in fēore ‘ever in life.’
My/mine also used to work this way, which you can see remnants of in phrases like mine eyes have seen the glory…
I’m referring to the so-called V2 (verb-second) word order. Students or speakers of German will be familiar with this phenomenon, where the verb likes to appear in “second position” in a sentence, although the way V2 order actually worked in Old English is much more complicated than how it works in German.
There is another possible explanation for nōn, which is that it comes from a different ne compound, one meaning ‘not at all.’ (Dunkel 2014). Dunkel, George E. (2014). “*nóh₁ ne ‘gar nicht’”, in Lexikon der indogermanischen Partikeln und Pronominalstämme.
The -g- may arise from the form nec, a shortened form of neque ‘and not’ (another *ne- compound) or it may just be there to prevent the two vowels e and ā from colliding.
Moreno Bernal, Jesús and Bautista Horcajada (1997). Sobre no y non en español medieval. Revista de filología románica 14(1): 345–361.
If you’re wondering about ne… pas, that’s too juicy of a story to talk about in a footnote. I’ll tell that tale on another occasion.
Here, ne is joined to the verb ‘to have,’ in yet another ne-compoud: næfde is ne hæfde ‘not had.’
This is called Jespersen’s Cycle, which we’ll discuss in more detail on another day, the same day we talk about ne… pas. It’s going to be great.
Fascinating. N and nn are also the negatives in Ancient Egyptian (Middle Egyptian Hieroglyphic) in which they negate the following sentence, akin to writing "it is not the case that”…