Week 8: Productivity and creativity

English Morphology

Fernanda Barrientos

2024-11-10

What we know so far about lexeme formation

  • If we want to derive nouns from adjectives, there is more than one option, morpheme-wise:
    • warm \(\rightarrow\) >warmth
    • modern \(\rightarrow\) >modernity
    • happy\(\rightarrow\) >happiness
  • Now think of any other adjective.
    • Form the noun.
    • Which suffix did you use, and why wasn’t it -th?

Productivity

  • Processes of lexeme formation that can be used by native speakers to form new lexemes are called productive
  • Those that can no longer be used by native speakers are unproductive, regardless of whether we recognize affixes within a word and know their meaning
  • However, productivity is not a categorical phenomenon:
    • Would you say that -ity and -ness are equally productive?
    • Is productivity measurable? 📏 🤔

Factors affecting productivity

  • Transparency: morphemes are easily “detachable” from the base and have a clear, unique meaning: -ness (but not -ity)
  • Frequency of base type: morphemes that attach to many different bases: -ness and -ity (but not -esque)
  • Usefulness: What is the ultimate goal of the derivation process?
    • Deriving nouns from adjectives 👍
    • Finding words for jobs performed by women ( -ess in “actress”, “stewardess”, etc.) 🙅‍♀️

Lexicalization

  • A derived word has been lexicalized when the meaning is no longer transparent
    • Meanings of complex words where we can see the way in which each morpheme adds meaning is said to be compositional
    • Lexicalized words have non-compositional meanings
      • Oddity ‘something odd’ (and not ´the quality of being odd’)
      • Locality ‘place’ (and not ´the quality of being local’)
      • Transmission ‘part of a car that transmits power from the engine to the wheels’ (but native speakers rarely think of this)

Restrictions on productivity

Here is what we know so far:

  • Categorial: To what syntactic category does a certain affix attach to? (verbs, nouns, adjectives…)
  • Phonological: Sometimes affixes will attach only to bases that fit certain phonological patterns.
    • -ize prefers nouns and adjectives with 2+ syllables, where the final syllable does not bear primary stress.
    • -en attaches only to bases that end in obstruents: darken, brighten, and deafen but *slimmen and *tallen
  • Semantic: The resulting meaning may also play a role
    • un- prefers bases that are not themselves negative in meaning: unlovely but not *unugly, unhappy but not *unsad.

Further restrictions on productivity

  • Etymological restrictions: Some affixes attach to particular subclasses of bases
    • -en attaches to native bases to form adjectives ( wooden, waxen but not *metalen or *carbonen).
    • -ic has the same function but attaches only to borrowings from French or Latin: parasitic, dramatic
  • Syntactic restrictions: Affixes may to bases with certain syntactic properties
    • -able generally prefers to attach to transitive verbs that can also be passivized: loveable, *snorable
  • Pragmatic restrictions: Some languages have affixes that attach only when the base denotes something pleasant/unpleasant
    • Dyirbal: -ginay ‘covered with’. But only for dirty things, so it would attach to the word for 💩, but not to the word for honey

Measuring productivity

  • Why is counting from the dictionary a bad idea?
    • Paradoxically, the more we find in a dictionary, the less productive it is
  • Baayen (1989): the less productive a word formation process is, the less transparent the words formed by those processes,
    • Thus, the less transparent the words, the higher their mean token frequency in a corpus.
  • We can count all tokens of all words formed with a particular affix, and then seeing how many of those words occur only once in the corpus (a hapax legomenon).
    • The ratio of hapaxes to all tokens tells us something about the probability of finding new forms using that affix:
      • The higher the probability of low frequency items, the likelier those items are to be new formations.

Baayen’s productivity formula

  • We can calculate the following:

\[ P = \frac{n1} {N} \]

  • Where:
    • N = total number of tokens
    • n1 = number of hapaxes
  • Examples from COCA:
    • -esque: 665 hapaxes/5738 tokens = .12 \(\rightarrow\) more productive
    • -dom: 257 hapaxes/64345 tokens = .004 \(\rightarrow\) less productive

Is productivity the same as creativity?

  • Very productive processes create very unremarkable words
    • Subliminality: “hm, new word for me, but nothing remarkable about it”
  • Morphological creativity uses unproductive processes or marginal lexeme formation processes, like blending or backformation.
    • The purpose is usually humorous:
      • Toebesity: ‘fat toes’
      • Situationship: ‘in an informal relationship’
      • Combobulated: ‘collected, no longer confused’ (from discombobulated)

Summary

  • Productivity is not a black-or-white phenomenon
  • Three main factors may impinge on productivity:
    • Frequency of base type
    • Transparency
    • Usefulness
  • Productivity may be restricted by etymological, pragmatic, and syntactic reasons
  • We can determine how productive a process is by looking at the hapax-to-token ratio
  • Productivity \(\neq\) creativity: the latter uses unproductive processes

Next week

  • We are having our lecture asynchronically!! (i.e. no lecture, but find materials on ILIAS)
  • Read Lieber, Ch. 4
  • Attend the tutorial
  • Finish the exercises