Common Morphological Patterns

Dr. Will Styler - LIGN 120

Today’s Plan

• How do we gloss sentences?

• How do languages vary in how they use morphology?

• What kinds of morphological machinery are common around the world?

Glossing Conventions

Translations often don’t give us enough detail

Soy de Denver (Spanish)

• ‘I’m from Denver’

Ngelay chadi (Mapuche)

• ‘There is no salt’

• Baker, M. (2002). On category asymmetries in derivational morphology. In Morphology 2000. John Benjamins Publishing.

Ka:háskiyakíre:réʔeriwa:ha. (Wichita)

• ‘Once upon a time they say there was a big village.’

• 😮

• Rood, D. S. (2002). Polysynthetic word formation. In Morphology 2000. John Benjamins Publishing.

We can do better!

• We can provide detailed glosses which relate the language forms to the translated meaning

• These are called interlinear glosses, and (often) follow the Leipzig Glossing Conventions

• The rules are defined here

Rough outline of the Leipzig Conventions

• 1. Word-to-word correspondence
  • Textual alignment from line to line
• 2. Set grammatical value abbreviations
  • Your book has a great list of them at the beginning
• 3. Separate morphemes within words using hyphens where possible
• 4. Use periods to mark non-separable morphological constituents within a word (when not segmentable)

Soy de Denver (Spanish)

• be.1SG.PRES from Denver

• ‘I’m from Denver’

Ngelay chadi (Mapuche)

Ka:háskiyakíre:réʔeriwa:ha. (Wichita)

Interlinear Glossing makes things much easier for linguists

• It allows quicker understanding of the components and elements of the sentence

  • With just translations, understanding requires analysis

• We see the morphology front-and-center

• This is a bit tricky in typesetting, but it works!

• Interlinear glossing is a great tool to use in describing data and words!

One specific element we need to see for LIGN 120

• Separate morphemes using ‘-’

• Separate within-morpheme meanings using ‘.’

• This is subtle, but crucial

• 

Mind you, some languages do more within-word morphology than others…

• Wait, hold on, that feels like a segue!

Do languages differ in how they use morphology?

(Yes.)

Morphological Type

A means of categorizing how languages put words together

• This is useful for comparing word-building strategies across languages

• Do words have lots of morphemes or few?

Do words have lots of morphemes or few?

• If words have few morphemes (like English or Chinese), they’re “isolating” or “analytic”

• If they have many (like Spanish or German), they’re “synthetic”

  • Languages where whole sentences are single words are sometimes called “polysynthetic”

English
IE:Germanic - All over

I did see the cat on the street, he’s cute.

I did see DEF cat on DEF street he-be.3sg cute.

• Only one morpheme in most words! This is Isolating!

Spanish
IE:Italic - All over

Esta escribiéndomelo

be.3sg write-GER-1sg.DAT-3sg.MASC

‘He/she is writing it to me.’

• Lots of morphemes per word! This is Synthetic!

Wichita
Caddoan - Oklahoma

Kiyakiicíwa:cé:hirʔasʔirhawi

‘There was the big buffalo lying there.’

• Uh… yeah. This is polysynthetic!

We also care about how easy the morphemes are to pick apart

In a synthetic language, do the morphemes blend together, or are they easy to pick apart?

• If the meanings have individual, easily separable forms, it’s “agglutinating”

• If the meanings merge inseparably into opaque forms, it’s “fusional”.

Turkish
Turkic - Turkey

• Those add together cleanly, and pick back apart. We’re agglutinating!

German
IE:Germanic - Germany

Donaudampfschifffahrtselektrizitätenhauptbetriebswerkbauunterbeamtengesellschaft

Donau-dampf-schiff-fahrts-elektrizitäten-haupt-betriebs-werk-bau-unter-beamten-gesellschaft

Danube-steam-ship-transport’s-electricities-head-operation’s-work-building-under-officials-association

• Look at those clean lines! Agglutinative!

Russian
IE:Slavic - Russian

Ti uhodila so mnoi

Ti u-hodi-la so mnoi

2sg.NOM away-go.IMPERF-past.FEM with 1sg.INST

“You left with me”

• Uh-oh! Those morphemes mean lots of things! This is Fusional!

Spanish
IE:Italic - All over

Durmiéramos

Durmiéramos

Sleep.2pl.SUBJ.PAST.IMPERFECT

• Whoa. All that is one morpheme? No easy lines? Fusion drives engaged!

It’s always a continuum

• English is isolating, but Mandarin is more isolating

  • Same with Synthetic: Nobody’s besting Wichita, but we don’t hold it against them.

• Most languages do some fusion and some agglutination

  • We call a language fusional or agglutinating when it becomes a habit

Morphological Type - Review!

• If a language has few morphemes per word, it’s isolating

• If the language has multiple morphemes per word, it’s synthetic

• If a synthetic language generally has easy-to-pick-apart morphemes, it’s agglutinating

• If a synthetic language generally has blended or combined morphemes, it’s fusional

… Which gets us into our next question

How are languages marking meaning, anyways?

Two main methods of doing meaning

• Concatenation

• Non-Concatentative Processes

Concatenative Morphology

Concatenation

Combining elements in a strictly linear order, end-to-end

• So, concatenative morphology adds morpemes one-after-the-other

  • fish-ing

  • in-conceiv-able

Common Methods of Concatentative Morphology

• Prefixes

• Suffixes

• Circumfixes

• Infixes

Prefixes

• Precedes the base

• in-conceivable

• u-hodi-t (away-go.IMPERF-INF) ‘To go Away’ (Russian)

Suffixes

• Follows the base

• cat-s

• u-hodi-la (away-go.IMPERF-PAST.FEM) ‘She went away’ (Russian)

Infixes

• Placed inside the base

• abso-f***ing-lutely

• Alabama Passive infix -l-

• 

Circumfixes

• Occur on both sides of the base, but not independently

  • For an affix to be a circumfix, both sides must be obligatory!

• German ge-X-t past participle (ge-spiel-t, from spielen)

• This Papuan Malay circumix ke-X-ang which nominalizes verbs

• 

In the Chukchi example below, Comitative forms are marked with a …

1. Prefix

2. Suffix

3. Infix

4. Circumfix

Remember, it’s not always so easy to pull things apart!

Phonology makes things less clear, sometimes!

• This Begak (Austronesian, Malaysia) infix -i- marking completive aspect

• 

• The infix is added, then the stem vowel is removed

• Can also neutralize contrasts

Sequences of affixes can make some of them feel infixy

• Un-thought-ful-ness

  • How do we know ‘-ful’ isn’t an infix?

Aside: Ordering of affixation

Recovering ordering of operation (and thus, meaning) is always tricky!

• Incorrect understanding of the ordering can lead to morphological ambiguity

• This appears well handled by hierarchy

• This is why some people create morphological trees, showing the components being modified

… but are we always adding something on?

Non-Concatenative Morphology

Sometimes, morphological operations change words without adding new chunks

• You can’t readily disassemble the pieces

• The meaning change is carried by modifications to the other morphemes

• … or nothing appears to change at all!

Common non-concatenative ways of adding chunks of meaning to words

• Conversion

• Stem Modification

• Reduplication

Conversion

• ‘Just change the usage to change the type’

• Sometimes, there are changes with stress shift

  • re’peat (V) -> ’repeat (N)

  • con’vict (V) -> ’convict (N)

• Sometimes, there are no changes

  • ticket (N, V)

  • drink (N, V)

• I computered all day yesterday.

Stem Modification

• Modify the stem to indicate a shift in morphological category or meaning

• This can take many forms

Stem Modification - Tone

• What’s the passive morpheme in Dinka?

Stem Modification - Vowel Modification (‘Ablaut’)

• What’s the 2SGF morpheme in Zway?

• Also, sing/sang

Stem Modification - Consonant Changes

• What’s the 2SGF morpheme in Zway?

Stem Modification - Other Processes

• Labialization
• Vowel Fronting/Backing/Rounding/ATR Changes
• Voicing Changes
• Lenition or Fortition
• Gemination or Lengthening
• Shortening
• Segment deletion
• Metathesis (swapping segments around)

Reduplication

• Copying all or part of the base

• 

• A big, BIG problem if you’re expecting different segments

… but in all of these cases, nothing’s added, just changed!

Wrapping Up

• Using reasonable glossing conventions makes life easier for everybody

• Languages have more or fewer morphemes per word, and meanings separate more or less cleanly

• Meaning can be expressed by concatening chunks of form in various ways

• … or by modifying the base in predictable and detectable ways

• Be prepared for all kinds of craziness when dealing with actual language data!

For Next Time

• We’ll have more data to work with!

Thank you!