Real Spellers

English Makes Sense!

A teacher new to structured word inquiry just sent me an email with this excellent question:

 I wanted to know if negate is a base? My students have been working on other words, but this one came up when doing a read aloud. My dictionary seems to indicate it is a base, but I am not sure and wanted some clarification before I start a lesson on it. Do you know?

Instead of providing my response in an email to this one teacher, I decided to share the question and answer here in Real Spellers. Since this question is from a teacher who has just had her very first exposure to this work in an hour session, it seemed apropriate to place this in the forum for beginners. As ever the point is not this particular word, but to get a sense of how to go about investigating such questions. So the first step is to remind ourselves of the principles by which any scientific inquiry should be guided:

  • Scientific inquiry seeks the deepest structures that account for the greatest number of cases.

     And the related principle...

  • Scientific inquiry seeks the most elegant solution -- that which invokes no unnecessary entities.

If you want more clarity on these foundational principles, I have posted information on these principles with examples from spelling investigations at this link

And now we should remind ourselves of the order of questions that we can use to guide any spelling question:


1)  Meaning: Although I generally know what the word negate means, I start by checking my sense of the meaning of this word by checking my Oxford "Mactionary." Here is what I find for the definitions of this word:

negate /nəˈɡeɪt/
verb [ with obj. ]    

   1 nullify; make ineffective: alcohol negates the effects of the drug.

   2 Logic & Grammar make (a clause, sentence, or proposition) negative in meaning.

   3 deny the existence of (something): negating the political nature of education


2)  Structure: This is the specific target of the teachers question. Asking if <negate> is a base is by definition a question about the structure of the word. One of the very first messages I communicate to children about words even to pre-school children who may not even know their alphabet is that every word they will ever encounter is either a base, or a base, or a base with something fixed to it. We know <negate> is a word, but, is it complex? Does it have a base and another element? 

Testing morphological analysis of a written word: The key orthographic principle we need to use to guide our investigation is that any hypothesized analysis of this word needs to pass both a "structure test" and a "meaning test". Here are the straight forward questions to resolve these "tests."

Structure test:  Does the hypothesized analysis have a coherent word sum? In other words, can I prove all elements in the word sum?

Meaning test: Do the hypothesized base and the resulting complex word share a common root originThe underlying denotation of a base can be traced back through the root origin, so we need to use etymological information (root) to prove our morphological (base) hypothesis.

(This use of the term root and base is crucial, and more precise than is often encountered in other sources. If you have not yet investigated this use of these terms, I highly recommend that you go to this link to consider the tutorial film on the terms "base" and "root" in that Real Spelling Gallery.)

With the principles of investigation established, we can now apply them to the question about the structure of <negate>.

First I look at this word and look to see if I recognize any potential affixes or bases. I suspect that the reason for this question in the first place is that the questioner noticed a final <ate> which is a familiar suffix <-ate> in many words (e.g., valid + ate --> validate; act + ive/ + ate --> activate) but not in others (e.g., room + mate --> roomate not *roomm + ate; hot + plate --> hotplate not *hotpl + ate).

Regardless of whether it is correct or not, I always recommend that word scientists make their hypotheses concrete with a word sum to help them think about it. So as soon as I have the question, "Is <-ate> a suffix, I create the following word sum. 

? neg + ate --> negate    

(I have recently begun the practice of signaling a hypothesized word sum with a question mark to signal that it is not a conclusion about the word structure, but a question.)

Because I know vowel suffixes like <-ate> replace final, single, silent <e>s (like in the structure of <activate> above) I now see two possible analyses:

? neg + ate --> negate    

? nege/ + ate --> negate

With these hypotheses I now have some raw data to use in an investigation. If either <neg> or <nege> is the base of <negate> I need to find evidence of this base element (a written base) in at least one other word. I have now reached the third step in my "Stuck on a Spelling?" chart.

3) Relatives: Here I am reminded to look for morphological relatives -- other words with a hpothesized base -- and to check that hypothesis by investigating that any such word really is connected in meaning by checking the root in a dictionary. To find morphological relatives, I am reminded to go to the Word Searcher

For this investigation, I decide to try <nege> first. If I find a word that is related in structure and meaning to <negate> I will be able to resolve which of these two analyses is preferrable. In fact as I write, before I do my search on the Word Searcher, I realize that this analysis will only work with a base <nege>! If I were to add a vowel suffix to <neg> I would be forsed to double the <g>! (neg(g) + ate --> *neggate). Since <negate> is spelled with a single <g>, I can now eliminate the first hypothesis.

Let's see what I find!

Just three matches...

Search Results for "nege"
(3 matches)

This is interesting data. All three can pass the structure test:

re + nege --> renege

re + nege/ + ed --> reneged

re + nege + s --> reneges

But do they pass the "meaning test"?

Off the top of my head, there is a tempting connection. My understanding of the word renege is to go back on a deal. which has a negative connotation. But a potential meaning connection off the top of my head is certainly not scientific evidence. To conclude that I have found evidence that these two words <negate> with hypothesized structure <nege/ + ate> and the word <renege> with the hypothesized structure <re + nege> pass the meaning test. Do these two words share a common root origin? 

I can now check a couple of references. The citation in my Oxford on my Mac gave me etymological information about the word <negate> that I did not share before. Here it is:

ORIGIN early 17th cent. ( sense 1 and sense3): from Latin negat- denied, from the verb negare .

Well, now I'm getting more excited about this hypothesis. I'm getting both meaning and structural hints supporting the trail of my thinking. The emphasis on the denotation "deny" ties closely to my understanding of the word renege. Although we saw that in the 3rd sense given in the definition of negate in my Oxford, I think my understanding of the word was still too tied to my sense of the word negative. While negative and deny clearly share a connotation this semantic link is not being confirmed with etymological evidence. I should be a good scientist and check another reference for the origin of <negate> as well. Off to the excellent!

Here's what I find...

negate (v.) Look up negate at
1795 (with an isolated use from 1620s), back-formation from negation, or else from Latin negatus, pp. of negare. Related: Negated;negatesnegating.

Whoa! There is even more here. Etymonline calls <negate> a back-formation. This is a facinating process of word formation that I know a bit about. Here is a definition from my "Mactionary"


    a word that is formed from an already existing word from which it appears to be a derivative, often by removal of a suffix (e.g., laze from lazy and edit from editor).• the process by which such words are formed.

So according to etymonline, the word negation was in usage before the word negate! But that doesn't make in any less of a word. And I see the same Latin root 'negare' that I got from my Oxford. Now to complete the meaning test. Does the word <renege> which is shares a plausible structural connection to <negate> actually share a common root origin. This time, I'll start with Etymonline, then to my Oxford:


renege (v.) Look up renege at
1540s, from Medieval Latin renegare, from Latin re-, here probably an intensive prefix, + negare "deny" (see deny). Related: Reneged;reneging.

Oxford on my Mac:

ORIGIN mid 16th cent. (in the sense desert (esp. a faith or a person)): from medieval Latin renegare, from Latin re- (expressing intensive force) + negare deny. 

We now have solid scientific evidence that the word <negate> is not a base, but is in fact complex with a base <nege> from the Latin root 'negare' carrying the underlying denotation 'deny' affixed to the common suffix <-ate> with the following word sum:

nege/ + ate --> negate

And this word sum is confirmed with the related word sum which shares a common root, and the familiar <re-> prefix:

re + nege --> renege


What I love about this investigation from this teacher who has had exactly one hour of PD on this topic is that she perceived a rich question about the word negate in conjunction with working with her students. By posing that question to me, I was forced to do what I call an "inquiry-led" version of structured word inquiry.  By this I mean, I did not know the answer when I began this investigation. All I had were the principles of scientific inquiry, and basic knowledge about how to test the structure and meaning of words with word sums and etymological references. I had never considered the structure of <negate> before. I had certainly never perceived the connection to the word <renege>. In fact, I'm sure if someon had asked me yesterday how to spell the word renege I would not have known. You'll notice that only now -- with the word sum and historical influences on the spelling established that the chart above directs me to consider the grapheme-phoneme correspondences. Without the meaning and structure of the word renege established, I would not have a good reason to explain the final, single, silent <e> at the end of this word. In fact there is a great deal of fascinating and impotant work about the orthographic representation of the phonology of this word that we could continue on if anyone out there is interested. I will just point to a fascinating discovery I made in the citation of this word in my Oxford:

renege |rɪˈneɪɡ|(also renegue )

Note that we have clear evidence that this word is pronounced witha  final /ɡ/ phoneme. And I also have evidnence that there are two attested spellings. I have a working hypothesis about the reason for the <u> in the second spelling, but I'll leave that for others to investigate if they wish. 

Before we move on to the phonological questions, however, I would encourage the teacher and students who posed this excellent question, and anyone else who is intersted to take the "Relatives" part of this process throug a richer process. We know there is a bound base <nege> from the Latin 'negare' for "deny". (We call a base a 'bound base' when it does not suface as a word on its own. A bound base only occurs in a word when it is bound to at least one other element.) We know that this base builds the words negate and renege. Why not look for more words that share this base element and thus expand our knowledge of the words that beling in this morphological family that share this common base element? 

So I encourage you to go off to the Word Searcher and type the letter sequence "neg" in the search engine. Remember, if you type in "nege" you will only get those words in which the bound base does not take a suffix at all, or a vowel suffix. By typing "neg" you are likely to get many false connections, but now you know how to test any word you encounter with a word sum, and an etymological reference. When you find a set of words that you are confident pass both the structure and meaning test, you can go to the mini-matrix maker and construct a matrix that you could share with everyone here on Real Spellers. (If you have trouble posting an image, just email me at This email address is being protected from spambots. You need JavaScript enabled to view it. and I'll help!) I'd love to see not only the original team of word scientists who identified and posed this question create such a matrix, but others around the world. 

If you have questions about the phonological lessons to be learned in this investigation, please post those as well. There is a community of scholars around the world ready to respond to such questions. I'm going to count on them to help bring precision to any of the language I have used in this post. It is not at all unlikely that there are places in this post where I have used misleading terminology or at least imprecise language or thinking. The only way I get better is by sharing my undertanding with others, an having others provide evidence of more precise ways to think about and communicate these ideas.

There is certainly nothing negative about making mistakes unless we deny that we make them!




Comments (0)

There are no comments posted here yet