The bifurcated nature of plural
Reconsidering evidence from English compounds
43rd Meeting of the Berkeley Linguistics Society
Jeffrey Punske
Southern Illinois University Carbondale
[email protected]
Scott Jackson
University of Maryland
[email protected]
I. Introduction
A longstanding assumption: regular plural morphology is generally banned on compound
modifiers (see Thomas-Flinders 1983, Kiparsky 1982, Hammond 1984 and Sproat 1985 among
others):
(1) Regular Plural Inflection in English Compounds
dog-lover
rat-chaser
log-cutter
hand-made
finger-bowl
coat-rack
three-twinkie lunch
*dogs-lover
*rats-chaser
*logs-cutter
*hands-made
*fingers-bowl
*coats-rack
*three-twinkies lunch
(Sproat 1985, p. 412)
In contrast, irregular plural morphology is has been assumed to be generally allowed:
(2) Irregular Plural Inflection in English Compounds
feet-first
mice-infested
teeth-marks
alumni-club
menfolk
(Sproat 1985, p. 413)
This observation lead frameworks like Lexical Phonology to divide the morphological system
into multiple strata:
(3) Regular morphology and compounding in Lexical Phonology (cf. Kiparsky 1982, as
in Sproat 1985).
Stratum I:
Irregular Inflection
Stratum II:
Compounding
Stratum III:
Regular Inflection
Punske and Jackson BLS 43
However, there are instances when regular plural morphology is available (or required) in
compounds:
(4) Plural Inflection on Pluralia Tantum Forms in English Compounds
pants-pocket
alms-giver
odds taking
(Siddiqi 2009, p. 57)
(5) Plural Inflection on Group Meanings in English Compounds
systems analyst
parks department
admissions committee
numbers racket
reservations deck
(Hammond 1984, p. 30)
(6) The left member of a compound must be unmarked for number,
unless the plural is interpreted collectively or idiosyncratically.
Sproat 1985: p. 420
 In this paper, we further develop Sproat’s constraint on compounds,
and demonstrate how both “exceptions” are derived straightforwardly
from independently motivated structures.
Main empirical claim:

there is no general ban on plurals (regular or irregular) for nominal modifiers in
compounds
Main theoretic claims:





Plurals are distributed over a bifurcated Number structure
o Lower number is either the head of nP or is contained within nP
 This is the domain of plurality
o Higher Num is outside of nP
 This is the domain of individuation
Compounds are derived via incorporation (Harley 2009; Jackson & Punske, 2013)
Incorporated modifiers must be a cyclic domain (phase)
nP is a cyclic domain, so the higher Num is never permitted in compounds
The phase-bounded nature of the lower nP permits it to instantiate idiosyncrasy, both
in morphological form (irregular) and in semantics
2
Punske and Jackson BLS 43

The ability for a compound to permit a modifier with a plural, individuated or either
interpretation is selection
o This follows from the complement relationship from Harley (2009)/ Jackson
and Punske (2013)
II. Data
Regular and irregular nouns can in fact occur as the compound modifier in either singular or
plural forms, though individual items may only occur in one or the other:
(7)
Singular
Plural
Regular
rat-catcher
dog-lover
*pant pocket
*odd maker
park department
number game
*rats-catcher
*dogs-lover
pants pocket
odds maker
parks department
numbers game
Irregular
woman chaser
toothbrush
*(wo)manfolk
*alumnus club
mouse infested
person mover
*women chaser
*teethbrush
(wo)menfolk
alumni club
mice infested
people mover
Critical distinctions between grammatical / ungrammatical ( or unavailable) forms:
(8)
Individuated
Collective
Regular
rat-catcher
dog-lover
park department
system analyst
parks department
systems analyst
Irregular
woman chaser
toothbrush
mouse infested
person mover
(wo)menfolk
alumni club
mice infested
people mover
(9) “Default” meaning: number game – a game that somehow involves numbers
(e.g., educational math game)
Idiosyncratic meaning: numbers game – involving gambling (alt. meaning of numbers)
(10) Pluria tantum: alms-giver, pants pocket, odds maker (arguably also idiosyncratic)
3
Punske and Jackson BLS 43
Summary: individuated plurals are not available in compounds, plurals with idiosyncratic
meanings (i.e., differing from singular) are available in compounds, as are pluria tantum forms
III. Analysis
Theoretical background
Compounds:
We follow Harley (2009) and Jackson and Punske (2013) in assuming that compounding
involves incorporation of the modifier:
(11)
nP
√P
n°
√i
√DRIVEi
nk
n°
√DRIVEi
er
drive
√TRUCKl nk drive
nP
nk
√TRUCKl

truck

truck
(Error! Reference source not
found. from Harley 2009: p. 136)
Key points:




Compound modifiers merge as nP complements to the root of the “head noun” of the
compound
Modifiers undergo incorporation into the head noun root
The nP of the compound modifier is a phase domain
Jackson & Punske (2013) use this structure in an analysis to derive so-called “compound
stress” from general phase-based stress rules, argued independently to apply at the
sentence level, with the nP phase of the compound modifier playing a key role
Number:
We follow Lecarme (2002), Borer (2005), Lowenstamm (2008), Wiltschko (2008), Alexiadou
(2011), Mathieu (2014), and Kramer (2016) in assuming that number is split between multiple
syntactic heads
(12)
Number domains adapted from Kramer 2016 (p. 534)
NumP
Num
= Domain of individuation
nP
n
= Domain of plural
√
4
Punske and Jackson BLS 43
For Kramer, Num and n may both host [+pl] features
Deriving idiosyncratic meaning:
If we assume that idioms/non-compositional meanings are phase-bound (cf. Svenonious 2005,
Harwood 2013, Harley and Stone 2013):
(13)
(14)
John kicked the bucket
John plays the numbers
Allomorphy:
We are largely going to side-step the ongoing questions of the derivation of allomorphy in DM
and related frameworks (see Embick 2010, Bobaljik 2012, Merchant 2015, among others)
The theory of allomorphy we assume most closely follows Embick (2010).
We assume that allomorphy is determined via syntactic adjacency and may be disrupted by
cyclic domains.
Thus is our structure, “low number” may be allomorphic while “high number” may not, which is
essentially identical to Kramer (2016)
Deriving Allomorphy
In our approach, the semantics of plural are also bifurcated, which separates us from approaches
like (Kramer 2016)
Low number: n is plural
High number: Num is individuation
We argue that English plural -s may be spelled-out in the low-position
Such an approach captures the difference between kind-plurals and individuated plurals.
(15)
Rats have tails.
But the English morpheme –s may also realize Num:
The structure here potentially allows us to explain allosemic forms:
(16)
person ~ people
(17)
person ~ persons
(18)
people ~ peoples
5
Punske and Jackson BLS 43
(An aside: one possibility is that English is prevented from realizing both plurals generally by a
haplology or Distictness restriction (c.f. Richards 2010)
Additional prediction:
This analysis predicts that individuation should never trigger allomorphy on a root or number
marker.
Analysis
 Only the nP, not the larger NumP, can incorporate as a compound modifier



Only non-individuating plurals are contained by nP, thus indivuating plurals cannot be
incorporated as compound modifiers
o This explains availability of collective-reading plurals (parks department)
Since the lower nP is a phase domain, it can also be a domain for idiosyncratic meanings
o This explains availability of idiosyncratic-meaning plurals (numbers game)
Both of these “exceptions” follow from the phase-based incorporation analysis of
compounds, argued for independently by Jackson & Punske (2013), based on stress facts
(19)
Illicit compound structure
nP
√P
n°
√
*NumP
rats
n°
√INFESTi
infest
ation
√INFESTi
infest
NumP
Numk
[indv]
s
nP
n°
[pl]

√P
√RAT
rat
6
Punske and Jackson BLS 43
(20) Well-formed compound structure
nP
√P
n°
√i
√LABi
nk
√Ratk
nk lab
rat
s
n°
√LABi

lab
nP
nk
√RATL
[pl]
s
rat
Because Num is larger than a phase it cannot be incorporated (following Harley 2009)
The complement relationship of the modifier allows us to capture otherwise difficult to capture
restrictions on modifiers through selection:
(21)
(22)
pants pocket ~ %pant pocket
*pants leg ~ pant leg
If this ability of have plural were a property of pluria-tantum-ness the ungrammaticality of pants
leg is unexpected.
However, √LEG selects for a non-plural (or possibly a +INDV- entailing non-plural) forcing the
result that is seen.
IV. Comparison with other approaches
Siddiqi (2009) offers a fusion-based account, wherein the Root and the NUM sometimes fuse
prior to compounding.
(23)
*Siddiqi (2009) Structure of *rats-infested
A
compounding v
√P
Num
rat
[n]
-ed
v
√INFEST
-s
7
Punske and Jackson BLS 43
(24)
*Siddiqi (2009) Structure of lice-infested
A
compounding
√P
v
-ed
v
√INFEST
lice
Node contains: √LICE, [n], [plural]
For Siddiqi’s analysis to work, forms with regular plural exponence need to be stored as whole
forms distinct from their unpluralized forms:
As Punske (2016) notes: “the form parks is stored wholly and separately from park so
that it can be inserted in the compound parks department”
Indeed, within Siddiqi’s analysis English contains two distinct forms parks
one that is park + s as is the product of regular morphological processes
another that is stored as parks so that is may occur in compounds such as parks
department
Concluding summary

There is no general ban on plurals in compound modifiers

We follow a phase-based incorporation analysis of compounds (Harley, 2009; Jackson &
Punske, 2013), where compound modifiers must be nPs

We follow numerous authors on assuming a bifurcated plural, such that individuation is
in a functional structure above the nP

It follows that only non-individuating plurals (regardless of regularity) can be compound
modifiers

Since nP is a phase domain, it also explains the availability of plural compound modifiers
when they have idiosyncratic meanings, by assuming these are nPs

This more elegantly and appropriately allows cases such as numbers in numbers game to
be “lexicalized” as whole forms, while simple non-individuated plurals like parks in
parks department can be derived normally from constituents √PARK and -s
8
Punske and Jackson BLS 43
References:
Alexiadou, Artemis. 2011. Plural mass nouns and the morpho-syntax of number. IN:
Proceedings of the 28th West Coast Conference on Formal Linguistics, ed. by Mary Byram
Washburn, Katherine McKinney-Bock, Erika Varis, Ann Sawyer, and Barbara Tomaszewicz.
Somerville, MA: Cascadilla Press. 33-41
Bobaljik, Jonathan. 2012. Universals in comparative morphology: Suppletion, superlatives, and
the structure of words. The MIT Press: Cambridge, MA.
Borer, Hagit. 2005. Structuring Sense. Oxford: Oxford University Press.
Embick, David. 2010. Localism versus globalism in morphology and phonology. The MIT Press:
Cambridge, MA.
Hammond, Michael. 1984. Level ordering, inflection and the Righthand Head Rule. MIT
Working Papers in Linguistics volume 7, ed. by Margaret Speas and Richard Sproat. Cambridge,
MA: MIT Press.
Harley, Heidi. 2009. Compounding in Distributed Morphology. IN: The Oxford Handbook of
Compounding, ed. by Rochelle Liber and Pavol Štekauer. Oxford: Oxford University Press. 129144.
Harley, Harley. and Megan Stone 2013. The 'No Agent Idioms' hypothesis. In Syntax and its
Limits, ed. by Raffaella Folli, Christina Sevdali, & Robert Truswell,. Oxford: Oxford University
Press.
Harwood, William. 2013. Being progressive is just a phase: dividing the functional hierarchy.
Universiteit Gent doctoral dissertation.
Jackson, Scott and Jeffrey Punske. 2013. Deriving English compound stress: insights from
Distributed Morphology and multiple spell-out. Linguistic Analysis 38: 243-274
Kiparsky, Paul. 1982. From Cyclic Phonology to Lexical Phonology. IN: The Structure of
Phonological Representations: Part I, ed. by Harry van der Hulst and Norval Smith. Dordrecht:
Foris. 131-175.
Kramer, Ruth. 2016. A split analysis of plurality: Number in Amharic. Linguistic Inquiry 47:
527-559.
Lecarme, Jacqueline. 2002. Gender “polarity:” theoretical aspects of Somali nominal
morphology. IN: Many Morphologies, ed. by Paul Boucher and Marc Plénat. Somerville,
Massachusetts: Cascadilla Press. 109-141
9
Punske and Jackson BLS 43
Lowenstamm, Jean. 2008. On little n, √, and types of nouns. IN: Sounds of Silence: Empty
Elements in Syntax and Phonology, ed. by Jutta Hartmann, Veronika Hegedűs, and Henk van
Riemsdijk. Amsterdam: Elsevier. 105-144.
Mathieu, Eric. 2014. Many a plural. IN: Weak referentiality, ed. by Ana Aguilar-Guevara, Bert
Le Bruyn and Joost Zwarts. Amsterdam: John Benjaimins Publishing Company. 157-182.
Merchant, Jason. 2015. How much context is enough? Two cases of span-conditioned stem
allomorphy. Linguistic Inquiry 46: 273-303.
Punske, Jeffrey. 2016. Compounding. Language and Linguistics Compass 10: 382-393.
Richards, Norvin. 2010. Uttering trees. The MIT Press: Cambridge, MA.
Siddiqi, Daniel. 2009. Syntax within the word: Economy, allomorphy, and argument selection in
Distributed Morphology. Amsterdam: John Benjaimins Publishing Company.
Sproat, Richard. 1985. On deriving the lexicon. Cambridge, MA: MIT dissertation
Svenonius, Peter. 2005. Extending the extension condition to discontinuous idioms. Linguistic
Variation Yearbook 5: 227-263
Thomas-Flinders, Tracy. 1983. Morphological structures. Los Angeles: UCLA dissertation.
Wiltschko, Martina. 2008. The syntax of non-inflectional plural marking. Natural Language and
Linguistic Theory 26: 639-694.
10