Sunday, February 5, 2012

Polysemous Prefixes over- and under- (Survey Results)

Thanks for participating in the recent survey! The prefixes under- and over- are among the top 20 most commonly used prefixes, according to the The American Heritage Word Frequency Book (Carroll, Davies, & Richman, 1971). According to that source, in the school texts examined, the prefix over- was the 8th most common, and the prefix under- was the 20th most common. More Words shows 406 words that begin with the letter string under and 1,681 words that begin with over.

The poll is closed. The results are shown below, and general discussion follows, including application to the Common Core State Standards and implications for instruction.

Post publication note: Most of the words in the survey are compound words. In the comments below, the point is made that over- and under- are not actually prefixes, but instead bases. Please see the comments for more elaboration on this excellent point.

Frankly, I had a tough time responding to this poll myself (yes, I know, I created it!!). Trying to be devious (do I get absolution for admitting that?), I included a word that begins with the prefix un-, yet looks like under- (underived). I also included a word that is misspelled (*overian, for ovarian). No one fell for my tricks (attributing the 2% to statistical error for underived).

The survey question that surprised me was #2: underage--undersized. I expected more people to view these two words as sharing the same prefix, with the same meaning 'not enough, insufficient' but only 72% did so (136 of 190 respondents). Not sure why. I could be missing something important. I was also surprised by overlearn--overboard (I view them as different senses of the prefix) and by overfed--overnight (different, I'd say). Some words are just too nuanced for me, as with underlying--undergoing. A colleague thought they do share the same meaning, where the prefix means 'beneath, under'. She said, "Any underlying cause rests beneath all the symptoms, and when you undergo surgery, you go under anesthesia, or under the knife." If we can get students to explain their thinking in such ways, we are hitting the jackpot.

Think Map, Visual Thesaurus
To help me understand one survey item, I turned to the Visual Thesaurus. The Think Map to the left, generated for the targeted word underhanded, helps explain why 32 respondents felt that underarm and underhanded shared the same sense of the prefix. One colleague said that underhanded and underarm are different, because underarm suggests something spatial, directional, and physical, like a baseball pitch, but underhanded suggests something hidden and covert (sneaky, as shown in the Think Map). However, another colleague disagreed, stating that her husband is a coach, and her first thought was that you can throw an underarm pitch or an underhanded pitch (also included in the Think Map). Context matters! Prior experiences with the word make a difference! Instructional discussions matter, too.

Rather than going over each survey item, let's discuss the various meanings of the prefixes and several instructional applications. (Even without knowing the "right answer" to the survey items, the thought process is beneficial. This is an exercise teachers might share with students, but using known words, and providing some context.)

Native English speakers and Dual Language Learners might assume that over- means 'on top of, above' and under- means 'beneath, below' (as in overcoat, undershirt) but that is not the only interpretation. Over- also suggests 'too much' and under also suggests 'insufficient, not enough' (as in overexcited and underexposed). Words that contain the prefixes over- and under-  are more likely to suggest the notion of 'too much' or 'not enough', compared to a spatial meaning.

These are broad interpretations of the two prefixes. However, a closer consideration reveals more than two meanings for each prefix. Michael Quinion, at, provides a brief summary of the various interpretations of the prefix over-.  He states, "Its meanings are rather variable and diffuse, and difficult to categorize."  (Yes, indeed!)

Quinion describes multiple ways the prefix over- is used to convey word meaning:
1) suggests 'something beyond what is usual or desirable, even excessively so' (overambitious, overcareful, overfull, overprecise)

2) suggests ‘utterly’ or ‘completely’ (overawed, overjoyed, overcome) [as in overcome with emotion]

3) suggests 'a spatial sense of something above or higher up' (overhang, overarching, overlook)

4) suggests a figurative sense of 'something that is superior' (overseer, overlord, oversight committee)

5) suggests something 'outer' (overcoat, overshoes)

6) suggests something 'extra' (overtime, overpay) [an oxymoron]

7) suggests 'a motion forward and down, and hence of inversion' (overturning, overbalance, overthrow, overboard)

8) suggests 'covering a surface' (overpaint, overgrowth).

Implications for instruction: I can see no reason why a typical language user needs to know all these nuanced senses of the prefix, but school children should know the most commonly used interpretations. Furthermore, they should learn to think about this type of word analysis. With so many nuanced meanings of the prefix over- (and the prefix under-) it is clear that we cannot rely on simply memorizing lists of morphemes and their meanings. We might also encourage children to think about how morphemes and words interact together, with context, to convey meaning. The value is in the critical thinking and the discussion that students engage in when working collaboratively through these types of questions. Tip: When asking students to sort meanings, an "other" category is useful, for those words that are difficult to classify, and an "unknown" category helps separate the others from the unknowns.

Why bother? Teaching children to think in these ways increases their level of morphological awareness. Readers who reflect on various aspects of language are more likely to comprehend words, phrases, and passages (see post on metalinguistic awareness). In addition, interpreting affixes is one of the learning goals outlined in the Common Core State Standards for English Language Arts.

Application to The Common Core State Standards
      Vocabulary Acquisition and Use

  • L. 1. 4 [First Grade] Use frequently occurring affixes as a clue to the meaning of a word.
  •  L.2.4 [Second Grade] Determine the meaning of the new word formed when a known prefix is added to a known word (e.g., happy/unhappy, tell/retell).
  • L. 3. 4 [Third Grade] Determine the meaning of the new word formed when a known affix is added to a known word (e.g., agreeable/disagreeable, comfortable/uncomfortable, care/careless, heat/preheat).
    Source: English Language Arts Standards (Language subsection)
Questionable Lesson Plan: I searched the Web and found a lesson plan for teaching the prefix over-. I won't name the site, but it was not related to the Common Core. According to the webpage, the lesson applies to Grades 2 and 3.  I was happy to see the lesson, but only one meaning of over- is taught. Here is an excerpt:
Step 4. Define the meaning of over-, as well as words containing the prefix over-.  Look at the list of words with the prefix over-. Who knows what over- means? Over- means “too much.” Look at overwork. Overwork means “to work too much.” When the prefix over- is added to work, it changes the meaning of the word. Can anyone tell us what overslept means? What about overdo?
There is no additional instruction, no mention of the other meanings of over-, but the good news is that this is the most frequently occurring sense of the prefix. Nonetheless, students --and teachers-- who try to interpret overnight, overseer, and overcoat with the notion of 'too much' will run into trouble (especially since there is no context provided).

Note: Vocabulogic readers have participated in a number of surveys on common affixes. For examples, see The Popular Prefix in- (Survey)  and The Slippery Suffix -er (Survey).


  1. Anonymous2/05/2012

    Susan, thanks for the wonderful article. I completely agree that there's great value in having discussions about the multiple meanings a single prefix can have. It's always interesting to see how thinking evolves and switches start flipping when they start realizing that one small part of a word can change it's meaning. Dedra

    1. Right, Dedra. It is fun to see the "switches start flipping!" (love that expression)

  2. They are not prefixes...they create compound words. A prefix is a morpheme that can't stand alone.

    1. First, as to the essential bound nature of affixes, the linguistic definition is unequivocal. Here is David Crystal’s Dictionary of Linguistics and Phonetics.

      affix(-ation, ing) the collective term for the types of formative that can be used only when added to another morpheme (root or stem), i.e. affixes are a type of ‘bound’ morpheme.

      It is, then, crystal-clear that if an element is an affix it is, by definition, bound. The necessary corollary, then, is that if an element can be free it can not be an affix.

      By definition.

      Furthermore, from such constructions such as ‘overly’ and ‘underling’ we also have the evidence of the ‘base’ (rather than ‘affix’) status of both ‘over’ and ‘under’.

      Hereunder is further evidence!

      The comment is made, “However, they are classified as prefixes by a variety of experts, because they tend to be placed at the beginning of a word, rather than at the end of a compound”

      But “expert” or not, what what sort of term is “tend” in the establishment of scientific evidence?
      Evidence has the absolute priority over opinion, even that of ‘experts’ - so here is just a small part of further evidence that counters the asserted PREfixity of the base element ‘over’.

      MoreOVER, even if one gets the allOVER impression from the turnOVER that there might be some carryOVER or crossOVER from linguistics to pedagogy, such a handOVER must not be allowed to become a rollOVER takeOVER.

      And soon other compounds with ‘over’ as their second component raise their heads to remind us of their presence : ‘flashover’, ‘leftover’, ‘pullover’, ‘passover’, ‘hangover’, ‘sleepover’, ‘voiceover’... That’s hardly a ‘tendency; it looks more like a productive characteristic!

      - we have ‘over’ as a base taking the suffix ‘-ly’ to produced its own derivation ‘overly’ (and, of course, cricket-players all over the world know its plural ‘overs’);

      - over is said to “tend” to be placed at the beginning of a word, while we cannot avoid noting that it has a similar “tendency” to be place at the end of a word;

      -it occurs freely as a preposition, an adverb and even a noun.

      As far as concerns the notion of prefixes becoming ‘freed’ to become words in their own right, we should first check the evidence of the diachronic semantics of such lemmas and, as a consequence, will need revisit the concept of the ‘clip’.

      Finally, hats off and three cheers to Vocabulogic’s encouragement to investigate the semantics of affixes - qualitatively different from that of base elements.

      Such an initiative can only be applauded, and readers will be pleased to know that such orthographic activities are already rooted in some learning communities. Here, for instance, is a comment made to a group of fifth graders who, while working with the bound base element ‘ject’ whose orthographic denotation is “throw”, were considering the prefix ‘sub-‘ in the construction ‘subject’.

      At one point, you were discussing the sense of the prefix ‘sub-‘. You need to be careful when you are discussing the “meaning” of affixes. While base elements do have orthographic denotation, affixes don’t necessarily have them. And even when they seem to (e.g., ‘re-’ certainly seems have the denotation “(back) again”) they may also, or even only, have broader ‘sense’ or ‘force’. For example, ‘re-’ can simply be an ‘intensifier’. So with ‘sub-’ ; it certainly often does carry a sort of denotation “under”, but also has senses of “turning upward, close to, up to, towards…”

    2. I totally enjoyed reading your rebuttal, Real Spelling. I especially chuckled at all the under- and over-ending compounds

      (but far more words begin with under and over, rather than end with them)

      Morphemes, by definition, carry meaning. They are the smallest elements of meaning in a word. The meaning may be nebulous or unclear at times, but it is still there. Yes?...I am not sure I quite understand you, on that point, but I do comprehend and accept your suggestion to use caution when discussing the meanings of affixes.

      Your comment got placed above my reply to CP Tillman, so this reads out of order, but at any rate, I concur with the point that a more correct term for under and under is base ( or combining form -- a term used in school texts and in teacher education), not affix.

      Yes, sub is a clip. We discussed clipping in a prior post. But it is still a word. My admittedly non-elaborated point, to CP, is that some flexibility is called for, when discussing language....and unlike some scholars, I view language as a soft science, with lots of tendencies, and not so many laws (discounting the anatomical aspects of speech production)

      But terminology certainly matters, even in soft science!

      I am going to "ponder all these things in my heart" (maybe I will have an epiphany).

      Again, thank you for contributing to the discussion!


  3. Yes, indeed, I agree with you; over and under are free-standing morphemes, also called combining forms. However, they are classified as prefixes by a variety of experts, because they tend to be placed at the beginning of a word, rather than at the end of a compound (Carroll et al., 1971; Quinion, 2012). Most combining forms--the elements that create compound words--are more interchangeable than that, in terms of position within the word. Affixes typically are bound morphemes, as you say. However, think about the prefixes sub- and dis-; both have been "freed" to become words in their own right, even as they are prefixes.

    Anyway, I agree with you; it would be more correct to call them combining forms.

    Thanks for pointing it out.


  4. This post and discussion brings up a number of excellent rich topics for investigation. The question that I would like to build on is the following:

    What is the reason to call “under” anything other than a base?

    I know that it is often described as a prefix in references. I had not heard it described as a combining form before, but I don’t doubt that it is.

    But what about this morpheme makes either of those descriptors more accurate than the term base?

    The morpheme “under” is accurately called a base element when it acts as a free word, when it acts as the base in a derivation (e.g., “underling”) or when acts as one of the bases in a compound word (e.g., “underage”).

    Unlike prefixes or combining forms, “under” is not a bound morpheme which must be bound to other element. This “boundedness” of prefixes and combining forms is signalled by the convention of a hyphen showing where such an element must be fixed.

    As scientists and teachers, what clarity do we bring to the spelling system by accepting two or three terms to describe “under” as compared to the conclusion that “under” has all the properties consistent with the definition of a base?

    Another question likely grows from mine. Is this a nitpicking question, or does it have value for teachers and reading researchers?

    I would like to suggest that scientific questioning of long held, widely-shared assumptions in educational resources provides extremely rich learning opportunities for teachers and students alike.

    I will use the next comment to share an example to explain what I mean by this.

  5. Here is the example I wanted to share. (Apologies for my inability to write short comments on Susan's blog!)

    Recently, Douglas Harper, the author of the excellent etymonline (on-line etymological reference) joined the discussion of a post by Gina Cooke, at her LEX (Linguist-Educator-Exchange) blog. He noted that it was asserted in a number of responses in that discussion that that the Oxford English Dictionary wrongly cites “-tion” as a suffix. At the time his reference followed the lead of Oxford on this citation, so he asked for evidence for our conclusions.

    It was pointed out in the follow-up discussion that Apple’s Oxford English Dictionary cite two examples of words using what it calls the “-tion” suffix. Those words are “relation” and “completion.” Analysis with word sums makes it evident that both of these words must use an “-ion” suffix.

    “-tion” hypothesis:
    *rela + tion → relation
    *comple + tion → completion

    “-ion” hypothesis
    relate/ + ion → relation
    complete/ + ion → completion.

    (Note: Vowel suffixes replace the single, silent “e” but consonant suffixes do not.)

    Until a word is found which can be analyzed to remove a morpheme spelled “tion” from a base or stem, there is no evidence of such a suffix.

    In response to this evidence, Mr. Harper changed his previous citation which was based on Oxford into the citation that is now on Etymonline. The new definition was taken directly from the wording of one of Gina’s responses in this discussion.

    syllable formed when the suffix -ion (from L. -io) is fixed to a base or to another suffix ending in -t or -te.”

    I see it as an important story that the author of an already excellent (free!) etymological reference heard an interesting argument that countered something that he had long assumed to be the case. He was curious to find whether there was evidence that demanded he revise his understanding. When that evidence provided a more coherent explanation than his previous assumption, he signalled his delight to have been taught something new, and to be able to make his reference that much more accurate for his readers.

    What an excellent science lesson this story has to offer children in classrooms. I have long used this scientific analysis to show children that like any reference, authoritative sources like the OED can make errors. As a result of this new story we can now also show children an example of an established authority seeking out and accepting the scientific evidence countering his own assumptions. In the context of a topic they understand, we can invite students to observe the process of scientific inquiry moving our collective knowledge forward.

    Go to this link to see a video of a Grade 5 class that regularly engages in this kind of scientific inquiry with words.

    All the way through this video you see students behaving like scientists investigating words. Watch at the end as a student uses their hard won knowledge to identify and reject a flaw in an educational resource that cites “-sion” as a suffix. Their everyday use of scientific inquiry arms them with the tools to avoid being mislead by references which have not yet questioned their own assumptions.

    This strikes me as a brilliant example of the kind of independent critical thinking and problem-solving that schools seek to facilitate -- and it was accomplished by studying spelling!

    If anyone can provide evidence that “under” is anything other than a base, I would be delighted to see it. If there is no such evidence, Vocabulogic has an opportunity to help refine the precision and accuracy of its wide readership on this issue.

    More important than the label we use for one element, this readership gets to test and refine their own understanding of the terms base element, prefix and combining form and bring that deepened understanding to countless classrooms.

  6. Hi Pete,

    As always, well said. Thank you for the astute comments.

    In this blog, I am speaking to an audience of grade school teachers, whose textbooks and teacher guides ( and the professional books published by leading vocabulary researchers from the field of education, like Steven Stahl, 1999) have, for decades, taken a different approach to terminology. I understand your point, that Vocabulogic could help refine the precision of the terminology used by educators, by referring to word elements as either an affix or a base, and not using terms like combining form, and not referring to over- and under- as prefixes, but as you know, that is what many school texts do, and what teachers are familiar with.

    That is not to suggest that teachers would not get your point...I fully expect that a number of our teacher-readers already were pondering whether over and under are elements used in compound words, and not affixes.

    This is a dilemma for me. The goal is not to upset the apple cart, in terms of existing curriculum, but rather to encourage insightful approaches to word study, approaches that will benefit school children, especially those children in the greatest need of accelerated vocab growth. I see the issue of terminology as a less vital issue, but still important. Certainly it will benefit everyone, teachers and children, if everyone used the same metalanguage.

    I'm just not sure I want Vocabulogic to accept the challenge you have presented. Must think more on it.

    At any rate, your comments will certainly be of great interest to all who want to learn and study more. Much appreciated!


  7. Why not upset the apple cart if that's how we get at the bad apples? Just because many people have done something a certain way for a long time doesn't mean that there isn't a better way.

    The "dilemma" that is facing anyone reading this is the same dilemma that has faced science forever. That is the dilemma of whether to pay attention to evidence, or to pay attention to tradition. The answer, for science, is clear. When a current theory fails to explain all of the available data, then it must give way when another theory can account for more of the data. That's how science works.

    The fact that "many experts" think something is not in and of itself evidence of anything at all. In linguistic history, at one point, many experts believed that phonemes were perfect, fixed physical entities; now, that theory has been displaced by the data that shows phonemes to be malleable psychological constructs. Many pedagogical experts once believed that some people, like dyslexics, were incapable of learning.

    When, as professionals, as experts, we put our work out there, when we make it public, we are taking the risk that someone might come along and show that we're wrong, at least in some aspects of our work. That's an occupational hazard, *especially* when we claim to be partaking in a scientific dialogue.

    Doug Harper of etymonline didn't just make the change to * because I am an "expert" and I said so. He made the change because he *saw* the evidence that I showed him, and he's an intellectually honest guy. He also emailed me as he was going through his site and manually changing all the entries with * to the structurally correct . He told me that themore entries he changed, the more convinced he was that it was right.

    I once heard Reid Lyon say at a symposium that "science isn't democratic." He's right. Let's not forget that before Niclas Koppernigk, or Copernicus published _De revolutionibus orbium coelestium_, many experts believed that the sun revolved around the earth.

  8. Not sure about any bad apples, per se.

    The reason I cite the expert sources is not because I think experts are infallible, but to explain how it came to pass that there are so many terms used in K-12 that are not used by linguists.

    How important is it, for linguists and K-12 language arts teachers to be on the same page?

    At any rate, I have inserted a note into the body of the post, alerting readers to these concerns.

    Copernicus would be proud.


  9. Rizalia Garcia4/12/2012

    Brilliant ideas and insights have been discussed here. As an INTESOL student, I am en tasked to explain the prefixes over and under and have been in so much confusion as to whether these two are prefixes or categorized as a compounding word. Moreover, I hope someone can help me what pejorative prefixes are especially the prefixes mis-mal- and pseudo.
    I'd love to hear any good responses.
    Thanks in advance

  10. Hi Rizalia,

    To summarize the comments above, over and under are bases or base words that appear in numerous compounds. They are not strictly prefixes, even though many researchers and curricular materials refer to them as such, and even though they tend to appear at the beginning of the compound ( but see examples where they appear last, in comment by Real Spelling ).

    If I were teaching today, using curriculum that classifies these as prefixes, I would facilitate a discussion or activity that helps students see the key points made in the post and the comments.

    I view the critical thinking as more valuable than the terminology...but terminology and meta- language can guide our thinking, so this does matter.

    Maybe someday we will offer a post about the morphemes mal- and mis-. For now, I suggest you visit and see what Michael Quinion has to say.

    Best wishes as you teach English around the world!

  11. Hi Rizalia,

    If you do go to Quinion's site, be very careful, as he is one of the many experts/sources who confuses bases with prefixes. For example, he offers 'orchido-' as a prefix, but it's actually a free base, 'orchid', plus a connector vowel, 'o'. That's just one error of many on his site.

    The only prefix you gave was 'mis-' which means 'wrong'. I wouldn't cal it pejorative, as it doesn't express contempt or disapproval. I'd simply call it negative.

    The element 'mal' is not a prefix; it's a base. We can see that it's a base in the compound word 'dismal' -- it can't be a prefix if it comes at the end of a word, can it?

    And the 'o' in 'pseudo' is a connector vowel; the evidence is in the word 'pseudepigraph'. In present-day English, 'pseud' has been clipped into a free-standing word, meaning 'false intellectual, sham.'


Comments are published after they are reviewed, to ensure they are not SPAM.