logo  ELT Concourse teacher training
Concourse 2

Morphology: the building blocks of language



What is morphology?

It is better to start with a definition of its subject matter.
A morpheme is the smallest meaningful unit of a language.
In this phrase, we have these 10 morphemes: the + small + est + mean + ing + ful + unit + of + a + language.
The term meaningful here does not refer to meaning standing alone (although it may) but to meaning in context or in combination with other words.
For example, a function word such as it carries no meaning without a context and co-text but in a sentence such as:
    I read the book and loved it
the word clearly carries the meaning of book.
And in
    John and Mary came to the show
the word and carries a meaning to the hearer / reader relating to combining subjects although the word standing entirely alone has no discernible meaning.
Both it and and are morphemes by this definition and so are I, read, the, book, love, -d, John, Mary, came, to and show.

Morphology is the study of how languages form words from smaller units.  All languages do this and they do it in a bewildering and fascinating number of ways.
We are concerned here with how English functions in this respect but we'll be referring to other languages as we go along because effective teaching in this area requires at least an outline knowledge of how your learners' language(s) function.


Identifying words

Before we can sensibly look at what words are made of, we need to be clear about what a word actually is.  This may seem a very simple question but it is actually rather difficult to define what a word is in any language.
Here's a definition from Google:

a single distinct meaningful element of speech or writing, used with others (or sometimes alone) to form a sentence and typically shown with a space on either side when written or printed

As we shall see, that is not an adequate definition for our purposes.  The definition is hedged with typically and sometimes and that is a sign that the definition is difficult.

think Task 1: Here are some traditional tests for words in a language.
Click on the table when you have filled in the right-hand column.

(Based on Crystal, 1987:91)

Of course, one ordinary way to identify a word is to look it up in a dictionary.  Unfortunately, that won't work, either.  If you go to an online dictionary for something like inderivability, you are very unlikely to find it but most native speakers of English would be happy to accept that it really is a word.
The other issue with dictionaries is that, in order to save space, they usually list words by the root or lemma.  For example, you will find repeat as a headword or lemma in a dictionary but repeated and repeating will normally not be listed separately unless the words are used with a different significance.  That doesn't mean they aren't words.

So what actually is a word?

For a morphologist, a word is any item which can be derived from the application of morphological rules.  This means that even something like computerhood, which you will certainly not find in a dictionary, is a word because it can be derived from the three morphemes that make it up, following normal conventions in English.  These are:
the verb
    + r to make the noun doer of the verb (or, with words that do not already end in 'e', 'er')
    + hood to make the noun signifying state of being a doer of the verb.


Identifying morphemes

think Task 2: Can you identify the morphemes in this list of lexemes?
Click on the eye open to reveal some comments when you have.

eye open
This is an easy one.  The lexeme is a single morpheme which stands alone.  We can add morphemes to it, of course, to make, e.g., sugary, sugariness etc.
It is a free morpheme which needs no other morpheme to make sense.
eye open
This is a compound formed from two free morphemes, class and room.  Both are nouns and they combine to make a new noun with a new but transparent meaning.  See the guide to compounding for more.
eye open
This is an example of a free morpheme combining with a bound morpheme to make a new word class.
We have taken the free morpheme,
ugly, and combined it with a bound morpheme suffix, ness, to make the noun from the adjective.  The suffix is a bound morpheme which cannot stand alone and carry meaning.  Because the morpheme is used to derive a new word, it is called a derivational morpheme.
(Do not worry about the spelling change.  That is simply one of the orthographic conventions in English and makes no difference to meaning.  Do not be concerned, either, with the fact that
ness can be a free morpheme meaning a headland.  That is an example of homonymy.)
eye open
This is an example of the same thing.  A free morpheme, teach, combining with a bound morpheme, er, to make a personal noun from the verbIn other words, to derive a new word class.  There is a second bound morpheme, s, which forms the plural in English, part of how nouns decline.
eye open
This, too, is an example of a free morpheme combining with a bound morpheme (fly and ing, respectively) but in this case we have a grammatical change, converting the base form of the verb into a participle or a gerund, depending on context.  This is an inflexional morpheme, determining how verbs conjugate.
eye open
This is slightly problematic.  Often, the bound morpheme prefix re implies the act of doing something again, as in rewrite, reinforce etc.  Here, however, analysing it that way would result in the morpheme peat which, as a verb, carries no meaning so cannot, by definition, be a morpheme in English.  It is, however, acceptable to analyse it that way because the verb derives from the Latin, via French, from re, [again] + petere [strive after; ask for].  It occurs again in, e.g., petition.
We have another inflexional morpheme here,
ed, which forms the regular past tense in English.
eye open
This, of course, can mean a person from Hamburg and that is an example of a free morpheme, called the root, combining with a bound morpheme to form the name for a person's origin.  Compare Londoner, Parisian etc.
In this case, however, the word probably derives from the name of a beef product from Hamburg and the morpheme
burger has taken on a life of its own as in cheeseburger, beefburger and veggieburger.
eye open
The function of the bound morpheme, dis, as a prefix is clear; it means the negative.  However, hidden in here is another Latin-derived meaning of re.  The word derives from re [back] + specere [look].
eye open
This is an example of how new words are formed from the general rules of a language's morphology.  We simply take the name or philosophy and add ite or ist.  Compare Trotskyite, Marxist, communist, sexist, Thatcherite, Benthamite etc.
eye open
This is a less common way to form new words called a blend.  Two words have been fused to form a third meaning.  Compare smog [from the free morphemes smoke + fog] or motel [from motor + hotel].
eye open
This is interesting because the bound morpheme, in, clearly carries a negative meaning.  Unfortunately, the second morpheme, ept, carries no meaning.  The word comes from Latin in- [not] + aptus [apt].
The term for this kind of morpheme is a bound base (sometimes a bound root).  It cannot exist alone but does not act as a prefix or suffix.  Other examples are the ver in verity, the doct in doctor and the dext in dexterity.

All the technical terms are in bold in this table.
Click here to take a test to see if you can remember what they mean.


Derivational morphemes

As we saw above, there are two ways that derivational morphemes added to lexemes can change them.

  1. We can change the word class or category of a word but leave the base meaning unchanged.  For example,
    • entertain (verb) → entertainment (noun)
    • entertain (verb) → entertaining (adjective)
    • elephant (noun) → elephantine (adjective)
  2. We can add morphemes which change the meaning but leave the word class unchanged.  For example,
    1. pleasantunpleasant (adjectives, opposites)
    2. doundo (verbs, reverse action)

Sometimes, we can add morphemes which change both the meaning and the word class.  For example,


Productivity of derivational morphemes

A central concern of morphology is to investigate how productive a derivational morpheme actually is.  This does not apply to inflexional morphemes because their role is generally fixed grammatically
How is it, for example, that the morpheme th can be attached to some adjectives to form the noun (warm → warmth, wide → width etc.) but is almost completely unproductive in forming new words in English?
We would not, for example, take the adjective windy and form a noun such as windith, preferring instead to choose the ness suffix and making windiness.
There are some interesting factors in play.



Some morphemes have an easy-to-see relationship between the form and meaning.  Here are some examples:

  1. It is easy to unpack a word like recordable by considering the suffix able and realising that it always produces an adjective from a transitive base verb and that it exists as a free morpheme meaning can.  You can quite easily make new adjectives like this from any number of verbs, providing they take an object.  The fact is that it is the only way English can form a new adjective from a verb.  We have, of course, the ible morpheme in words like flexible, audible, comprehensible and so on but if you try to form an adjective from unusual or phrasal verbs you will almost always opt for able.  Try it with:
        prune, devolve, explode, derange, disturb, put off, turn down
    The morpheme ible is, therefore, almost completely unproductive.
  2. Another example is the ways in English that nouns are formed from adjectives.  The two most common are:
        ness: happy → happiness, sad → sadness, great → greatness etc.
        ity: insane → insanity, absurd → absurdity, acid → acidity etc.
    However, if you try to form new nouns from these adjectives:
        blue-green, airy, snowy, freezing, wet
    You will almost always opt for ness, indicating that ness is more productive than ity in this function.
    This is not to say that ity is unproductive in the way that th is, but that it is less productive.  It can be very productive with adjectives ending in the able morpheme.  Try, e.g.,
        openable, disturbable, derivable, pickupable
    and ity seems to be the form of choice although it is certainly possible to form words like openableness.
  3. Other derivational morphemes are also less productive than others.  We can, e.g., form the doer of a verb by attaching the morpheme ant as in
        inhabit → inhabitant
        assail → assailant
        claim → claimant

    but this suffix is far less frequently used for this function than the ubiquitous er suffix.  Try forming the doer of the action from these verbs and you'll see how much more productive the er morpheme is:
        disturb, peruse, demand, accommodate
    The suffix ist is even less productive in this sense although typist and telephonist exist.
    Just as we saw for the able/ible distinction, the suffix or is now almost completely unproductive and is frozen into words like actor, transgressor and doctor.

Frequency and usefulness

Some suffixes are simply too constrained in the number of bases they can be attached to to make them very productive.
We can make an adverb from some nouns with the suffixes wise and wards as in, e.g.,
    crabwise, clockwise, northwards, citywards etc.
but the number of possibilities is very limited because of the infrequency of the resulting adverbs and the small usefulness the concepts have.

However, the adverb forming ly is hugely productive in its ability to produce adverbs from adjectives because the results are both frequent and useful.  It can also attach itself to the barely limited number of participle adjectives in English.  So we can form, for example:
    stunningly, interestingly, swingingly, boringly, understandingly, reassuringly
and so on.  We can even make up new adverbs on the spur of the moment and be understood, for example:
    She spoke praisingly.


Other constraints

some morphemes behave along phonological lines.  For example:
The verb forming ize/ise usually attaches to multisyllabic nouns and adjectives if the stress is not on the final syllable.  So we get, e.g.:
    random → randomize
    real → realize
    apology → apologize
Another verb-making suffix, en, attaches itself to single syllable adjectives but only if they end in certain phonemes (stops, and fricatives):
    deep → deepen
    light → lighten
    broad → broaden
    deaf → deafen
but not
    clear → *clearen
    high → *highen
in fact, we insert a 't' to conform with the phonological rule and get heighten.
This applies to prefixes, too, as we see with the choice often of im rather than in before /m/ and /p/ and ir before /r/.
The root of a word will often determine what derivational morpheme is possible.
For example, the adjective-forming suffix ic will not attach to Anglo-Saxon bases but en will so we get:
    vitriolic, metallic, dramatic
    woollen, leaden, ashen
The prefixes un, im, in etc. do not attach to words with negative connotations so we can't have:
    *unvile, *impernicious, *undoleful, *unhelpless
etc. but we can have
    unlovely, unappreciated, improper, unhelpful

Derivational morphemes are dealt with in greater detail in the guide to word formation.


Teaching implications of productivity and other constraints

think Task 3: Review the sections above on aspects of productivity and the constraints on word formation with derivational morphemes and consider for a moment what the implications are for teaching English.  Consider, too, production vs. comprehension.
Then click here.


inflexional morphemes

In English we can denote a number of grammatical constructions by using inflexional morphemes.  We usually do this by changing the ending but, irregularly, we can also change the internal characteristic of the lexeme.  These include:

  1. Tense:
    We saw above that the -ed/-d morpheme signals past tense but it also signals a past participle for regular verbs:
    and for some irregular verbs the past participle is also marked by a morpheme addition
    Internal changes are also morphemic but what usually occurs is that one bound morpheme is altered and becomes a slightly different bound morpheme as in, e.g.
        speak → spoke
        hold → held

    English is unusual among European languages in having only one past-tense form to apply to all persons.  Most other languages in this family use different endings for most of the persons and number.
    It is also unusual in often having the same form for the past tense and the past participle.  Other languages, such as Greek, distinguish the forms.
  2. Aspect:
    English signals progressive and other aspects by changes to the verb:
        go, going, gone
    Again, English is very limited in this respect having only one form of the past participle and only one of the -ing form.  The -ing form, incidentally, has no irregularities.  None.
  3. Person:
    Apart from the truly irregular verb be, English only signals person with the third-person -s morpheme (and then only in the singular):
    By European language standards, this is a very limited range.  In other European languages, for example, the translation works like this:
    English French German Polish Estonian
    I go
    you go
    she goes
    we go
    they go
    je vais
    tu vas
    elle va
    nous allons
    ils vont
    ich gehe
    du gehst / Sie gehen
    sie geht
    wir gehen
    sie gehen
    ty idź
    ona idzie
    ma lähen
    sa lähed
    ta läheb
    me läheme
    nad lähevad
    This is not universal and, for example, all three Scandinavian languages (Norwegian, Swedish and Danish) have the same form for all the verbs in all persons (går).
    For this reason, many languages, such as Greek and Italian, routinely drop the pronoun because the person is signalled by the verb ending.
  4. Number:
    Although there are some irregularities, English uses the s/es morpheme to signal the plural.
    Almost all nouns function this way in English which is a simple system in comparison to many other languages (and more complex than many which do not signal the difference between singular and plural at all).
  5. Adjectives:
    English uses morphemic addition to change some adjectives from absolute to comparative or superlative:
    The morpheme most can also act as a suffix carrying the same meaning in words like nethermost, uppermost, hindmost etc.  The morpheme more can't do that and the suffix most is now unproductive.
    The added complication in English is that some adjectives cannot take the inflexion at all so we do not allow, e.g., boredest or beautifuller.  The rules for when we inflect and when we use the periphrastic form are quite complex, in fact.
    Other languages are often simpler in this respect.  Most Italic languages, French, Spanish, Romanian, Portuguese etc. employ the periphrastic form and some, such as German, usually employ only the inflected form (although the periphrastic form is available), others, Scandinavian languages and some Slavic languages, for example, work a little like English.
    Adjectives in English remain unchanged, whatever case, gender and number they are associated with.  Other languages will often inflect the adjective to show case, gender and number.
  6. Case:
    Many languages, which take advantage of morphemes to signal case (nominative, accusative, dative, genitive, locative, ergative etc.) alter the nouns and other elements to show this.  English generally does not do this with lexical words but has a slightly complex pronoun and determiner system which sometimes radically alters a morpheme to show case and sometimes simply adjusts it slightly:
        I, me, my, mine
        you, your, yours
        she, her, hers
        who, whom
    English also deploys the 's morpheme for the genitive:
        John's, the dog's, the ship's, the government's
    The fact that English has two genitive structures (the government's policy vs. the policy of the government, for example) is an added complication which leads to a good deal of unnatural language in learners and some error.

Old English had many more (and more complex) inflexions than Modern English has retained.  The loss of inflexion is one of the most important changes to have occurred in the language.
Many other languages, such as German, Polish, Finnish, and French deploy a much wider range of suffixes to denote number, person, case and gender.  Other languages, such as the Chinese languages are even more limited than English in their use of inflexions and suffixation.

Related guides
semantics for more on meaning
types of languages for a guide to how other languages do things differently
compounding for how this type of morpheme manipulation works
word formation for more on how morphemes combine in English

There's rather a lot in this area which is a much-researched field and much of it is not relevant to teaching.
Crystal, D, 1987, The Cambridge Encyclopedia of Language, Cambridge: Cambridge University Press
Lieber, R, 2009, Introducing Morphology, Cambridge: Cambridge University Press