Download Reflexivity and Intensification in Middle English

Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
1
9.3.01
Reflexivity and Intensification in Middle English1
Claudia Lange, TU Dresden
0.
Introduction
One of the many remarkable changes the syntactic structure of English underwent
during its history concerns the way reflexivity is expressed, as the following examples
indicate2:
(1) He cwæδ: δine stemne ic gehire, leof, on neorxnawange, & ic ondræde me for δam δe ic
eom nacod, & ic behyde me ‘And he said, I heard thy voice in the garden, and I was afraid,
because I was naked, and I hid myself’ (ÆGen, quoted after Keenan 1996:6)
(2) Christus se dedit pro nobis Crist sealde hyne sylfne for us (ÆGramm (96.5))
(3) Hie gewendon heom to ðam cynge (‘They turned them(selves) to that king’) (Chron an.
1046)
(4) On cros godd boght ur saul liues, þar-on he gaf him-seluen ranscun ‚on the cross God
bought life for our souls, thereon he gave himself as remission‘ (Cursor Mundi 1244, quoted
after Peitsara 1997:299)
(5) Halde þe wel payed ‚hold yourself well paid‘ (Gawain 2341, quoted after van Gelderen
1999:205)
(6) He shall repente hym (Paston Letters, 143 (1452), quoted after van Gelderen 1999:209)
In Old and Middle English, the simple personal pronoun was normally used to express
the reflexive relation, as in (1) and (5). Only ocasionally and in specific contexts, as I
will show below, was the intensifier SELF, a free form, added as in (2). This pattern
continued to be used throughout the Middle English and into the Early Modern English
period before it gave way to the form of reflexive marking with pronoun+ SELF. Further,
constructions such as (3) and (6) with a non-argument or pleonastic reflexive pronoun
died out, (6) representing a set expression with a French loan when the pattern was
already quite rare at that stage.
Reflexivity in the most general sense can be understood as the marking of coreference
of subject and object; reflexive pronouns3 occur syntactically as objects of verbs
indicating coreference with the subject NP (a nominal antecedent in the same clause)
or as complements of prepositional phrases; they are arguments of the verb. For OE
1
This study is part of my Ph.D. which in turn is part of the research done within the DFG-project
„Typologische Untersuchungen zu den emphatischen Reflexiva, zur Reflexivität und zu den
Fokuspartikeln (Schwerpunktprogramm „Sprachtypologie“)“.
2 Unless otherwise indicated, OE examples are quoted after Venezky et al. (1980), A Microfiche
Concordance to Old English. Toronto.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
2
9.3.01
and ME, the definition of reflexivity needs to be extended to cover what has been
called ‘pleonastic reflexive dative’ (Mitchell 1985:113), ‘non-argument’ or ‘non-theta’
reflexive (Keenan 1996) as in (3) above.
In Modern English, the expression corresponding to German selbst is identical in form
with the reflexive pronoun, though not in distribution. The meaning of x-SELF in an
example like The queen herSELF declared the bazaar open is clearly not reflexive and,
for lack of a better term, was commonly characterised as „emphatic“; a term which is,
as has repeatedly been pointed out, not only vague but misleading: „it is necessary to
distinguish the kind of emphasis signalled by -self from other types of emphasis
(signalled by, for example, repetition, or constructions like topicalisation or clefting)“
(Kemmer 1995:57); a better term is ‘intensifier’.
Today, English is unique among the Germanic languages in employing a complex
expression, made up of personal pronoun+SELF, as reflexive marker also serving as
intensifier. Not having a
reflexive marker is a situation not uncommon within the
language family, also found, for example, in Frisian, Old Dutch, and Afrikaans (König &
van der Auwera 1994:XI). In what follows, I will try to trace some aspects of that
development that set English apart from other Germanic languages; I will also consider
whether that change was linked to the major structural changes that took place in the
period under discussion.
The organisation of this paper is as follows. In the first section, I will present the
analysis of intensifying x-SELF in Modern English that forms the basis for the following
foray into Middle English. The second section is devoted to sketching the development
of the complex reflexive and intensifier in Middle English, while section three
summarizes my main points and reviews further interesting questions I have not
touched upon (yet).
1.
X-SELF: a synchronic analysis4
The fact that Modern English does not formally distinguish intensifier and reflexive,
whereas Old English does, is clearly in need of explanation. SELF underwent a gradual
process of losing its independence as a lexical item and becoming fused to the oblique
In the following, ‚simple reflexive‘ will refer to the plain personal pronoun used reflexively and
satisfying an argument position of the verb, ‚compound reflexive‘ to pronoun+SELF when used
reflexively.
4 The analysis presented here relies heavily on König & Siemund (1996, 1998, 1999) and
Siemund (1997). I will not touch upon alternative analyses that try to accomodate the
peculiarities of x-SELF within the framework of Binding Theory, such as Reinhart & Reuland
(1993). See Siemund (1997) for the argument that the Binding Principles do not tell the whole
story about the distribution of reflexive anaphors and pronouns, neither in Modern English nor in
the earlier stages of the language; neither do they have much to say about x-SELF as intensifier.
3
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
3
9.3.01
personal pronoun, giving rise to the polysemy of reflexive marker and intensifier noted
above:
Old English
Modern English
self, sylf, seolf
free form (adjective)
intensifier (postnominal)
identifier/marker of reference (attributive)
x-self
bound morpheme
element of intensifier
element of reflexive pronoun
noun5
According to König (1991), intensifiers in general can be analysed as a special case of
scalar additive particles. Siemund (1997:11ff) argues for subsuming intensifiers under
focus particles despite the obvious incongruities between the two word classes, mainly
because of their parallel function: both interact with the focus structure of a sentence:
„All sentences containing intensifiers divide into a focused or highlighted part and a
background part, i.e. intensifiers structure propositions. Moreover, intensifiers
always evoke alternatives to the focused constituent“ (Siemund 1997).
Further, both focus particles and SELF have positional variability in common, and both
can occur more than once in a sentence. Unlike invariant, unstressed focus particles,
however, x-SELF displays agreement with its focus and carries stress. x-SELF is also
restricted to nominal foci, while particles may occur with a large variety of word
classes. Finally. x-SELF generally occurs in a position behind its focus, particles appear
before their focus.
focus
agreement
(pro-) nominal
intensifier
constituent
ModE distinguishes three separate uses of pronoun+SELF, apart from its use as
reflexive anaphor indicating co-reference in a local domain, being referentially
dependent on some preceding N. They all behave like adjuncts, either to some NP if
adnominal or to the VP if adverbal.
Siemund assigns all three uses of intensifiers to a common denominator, namely „the
ability to structure sets into a central element on the one hand and peripheral elements
on the other.“
In the taxonomy to follow, the first term refers to syntactic, the second to semantic
properties of pronoun+SELF in this particular usage.
5
The question of when exactly SELF became a noun or could be used as a noun is difficult to
answer, see Mitchell (1979).
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
4
9.3.01
1.1.
Adnominal, centralising SELF
(7) The minister HIMSELF will receive us
Adnominal HIMSELF follows its focus and bears stress. There are hardly any syntactic
or semantic restrictions on the focus. The NP to which x-SELF is right-adjacent may be
subject, object, or complement of a preposition. It may be a proper noun, a common
noun or a pronoun, with restrictions applying to pronominal head NPs, as the
discussion will show.
The contribution adnominal HIMSELF makes to the meaning of a sentence is to mark
the focus as central in relation to possible alternatives: the Minister HIMSELF -rather
than some lesser beings around him- will receive us. The focus accompanied by
pronoun+SELF forms the centre among similar entities that are assigned to the
periphery. Negation does not affect this meaning of SELF.
This meaning of x-SELF is likely to form the historically prior as well as the core
meaning of the focus particle x-SELF. At least x-SELF in adverbial-inclusive use is
closely connected with the adnominal use: I'm a bit short MYSELF = I MYSELF am a
bit short, TOO.
More precisely, the meaning of adnominal x-SELF can be stated as follows (Siemund
1997:192):
Adnominal intensifiers structure a set into a central element X and peripheral
elements Y.
a. X has a higher position than Y in a hierarchy
b. X is more significant than Y in a specific situation
c. Y is defined in terms of X
d. X is the centre of perspective (logophoricity).
The relation between the central element X and the peripheral elements Y may take
the form of one of the four specific relations listed above, as illustrated by the following
examples6:
a.
The Pope himSELF does not know what to do.
b.
Most of the passengers suffered light injuries. The driver himSELF was killed.
c.
Adam’s wife was picking apples, Adam himSELF was peeling them.
d.
He was not particularly tall, a little taller than Jemima herSELF perhaps. (A. Fraser, A
Splash of Red)
Following Baker (1995:80ff.), Siemund differentiates situational and organisational
centres as possible foci of adnominal SELF7. Organisational centres are centres in their
own right and not in need of further justification, they occupy their position due to
extralinguistic factors and independent of the current context of discourse. Situational
centres, on the other hand, receive their prominent role within and from a specific
context and constellation.
6
source: DFG- project website (http://www.philologie.fu-berlin.de/; link to ‚Forschung‘)
7
Baker uses the terms discourse-internal vs. discourse-external justification.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
5
9.3.01
Adnominal SELF places hardly any selectional restrictions on its focus; the NP it
intensifies may denote human or non-human referents.
1.2.
Adverbal SELF
Like adnominal intensifiers, adverbal x-SELF in both its inclusive and exclusive use is
always in association with an NP, though not a member of it. Adverbal x-SELF never
occurs adjacent to the NP with which it shows agreement and is best analysed as
belonging to the VP, or more precisely, as a VP-adjunct or an endocentric expansion
of the VP. It mainly occurs in typical adverb positions, e.g. sentence-final and between
auxiliary and main verb. Siemund demonstrates, however, that they do hardly behave
like other adverbs, and suggests the term ‘focusing adverb’ as the least unfitting label
because of their association with an NP, their carrying stress and their semantic
property of evoking alternatives (78).
1.2.1. Adverbal, inclusive
SELF
(8) Could you lend me ten pounds? - I'm sorry, but I am a bit short MYSELF.
The utterance containing the intensifier could be paraphrased with additive focus
particles as I am a bit short, too or I am also a bit short. With this utterance, the
speaker is in fact including herself among the set of contextually given possible
alternatives to her own person, therefore the label ‘inclusive’ to specify the semantics
of this particular use of SELF.
The focus of adverbal x-SELF has to be the subject denoting a human referent; in terms
of thematic roles, the focus is an EXPERIENCER. x-SELF appears as part of the VP
rather than the NP headed by the focus.
Again x-SELF places special emphasis on its focus, but only within a narrowly defined
context: x-SELF assigns prominence to the focus compared with the periphery which is
given by the immediate context.
1.2.2. Adverbal, exclusive SELF
(9) The girls painted the flat THEMSELVES (=on their own, without help)
In this case, the focus has to be an animate, agent subject, but not necessarily human.
The action denoted by the VP must be capable of being carried out by other agents as
well, otherwise an exclusive interpretation of the focus particle is blocked: In the phrase
Paul is snoring himSELF the action of snoring cannot be assigned alternatively to
somebody else; therefore the focus particle bears the inclusive meaning (Paul also
snores).
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
6
9.3.01
The analysis above, then, leaves us with the following picture of the meanings of
SELF8:
x-SELF
reflexive pronoun
adnominal, centralising
intensifier
adverbal, inclusive adverbal, exclusive
The questions now are which of these meanings were already present in the older
stages of the language, why the original monomorphemic intensifier disappeared and
why the complex reflexive marker consisting of pronoun+SELF became obligatory.
In the next section, I will show that in Middle English, the semantics of intensifying SELF
still determines its use when it occurs adjacent to a reflexive pronoun; the compound
reflexive becomes obligatory only later. The rise of the new form is linked to another
process, namely the realignment of the middle domain: the reflexive constructions with
a pleonastic pronoun, which can be interpreted as a kind of middle, die out, so that the
only option to express reflexivity is marking with pronoun+SELF.
3.
Middle English
Most studies concerned with Middle English mention the transitional quality of the
language of that period, the profusion of forms and possibilities available which make a
tight analysis difficult, and I will make no exception, but will then proceed to my topic
and rely on Fischer (1992:208), who states that
“the major syntactical changes in the Middle English period do not find their origin in
dialectal variants, but are a result of the morphological developments discussed
above. These are common to all Middle English dialects. It is true that individual
dialects may have undergone these changes at different times, but the ultimate
results do not essentially differ."
Before I look into the development in ME, I will sketch how reflexives and intensifiers
were distributed in OE, for two reasons: the processes that I am looking at originate in
late Old English and range over the entire Middle English period. Secondly, tracing the
source meaning of an item undergoing grammaticalization is important for establishing
in how far the original semantic substance determined its later distribution once it is no
longer an independent lexical item.
German has a further use of the intensifier that can be paraphrased by ‘even’, e.g. Das habe
selbst ICH verstanden.
8
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
7
9.3.01
3.1.
SELF
in Old English
As already noted, OE used the simple personal pronoun to express both conjoint and
disjoint reference; a sentence like Judasi hinei,j aheng is therefore ambiguous between
two interpretations: Judas might have hanged himself or somebody else9.
Mitchell (1985: 187ff.) classifies OE SELF as ‘pronoun/adjective’ and further categorizes
it as an ‘indefinite’ belonging to the subgroup of ‘words marking identity and the
contrary’ together with quantifiers like eall ‘all’, ilca ‘the same’ etc. SELF may precede or
follow its head N just as other adjectives (though preposing was more common for
adjectives) do, and it follows the usual rules of adjectival inflection, alternating between
weak (indefinite) when the NP is introduced by a definite article or demonstrative and
strong (definite) if not. Unlike adjectives, SELF cannot be compared, and it „can be
used both dependently and independently“ (103), that is either with a noun or instead
of a noun; SELF alone in nominative case occurs when a pronoun subject is left
unexpressed in paratactic sentence structures.
When used like a pronoun, SELF is usually declined strong, with some exceptions10.
Usually, SELF immediately follows its focus.
The loss of inflectional morphology which affected SELF as well was already apparent
in late Old English, where forms like sylfan or selven occur. Self(en), selue(n) occur in
ME without being indicative of case or number distinctions, with {-s}-plurals becoming
the rule around 1530 (Peitsara 1997:283). The personal pronoun paradigm lost the
distinction between accusative and dative, the dative pronoun him taking over as
oblique form11.
3.2.
Reflexivity
Information about the development of the reflexive in the standard ME handbooks is
scarce; Mustanoja (1960 429ff.) simply notes that „since OE there has been a steady
drift from the reflexive towards the intransitive form“, a tendency he attributes to
phonological change,12 French influence13 and mainly to „the inherent aversion of
English speakers to the reflexive form.“
Note that ambiguity arises only in the third person.
the two exceptions referred to by Mitchell are weak selfa following a noun or pronoun in
nominative singular, e.g. BlHom 13.26. he sylfa, WHom 11.107 God sylfa, and uninflected self
after nominative plural, e.g. Or 42.23. þa men sylf, Or 96.6. hie self
11 The ultimate loss of the accusative-dative distinction in the third person singular did not
proceed in a linear fashion. Kentish kept up the distinction well into the fourteenth century, while
the Peterborough Chronicle, written in the mid-twelfth century in an East Midland dialect, already
displays dative-accusative syncretism to a large extent (Howe 1996:134). For the West Midland
dialect, the collapsing of forms can be dated to the first half of the 13 th century (ibid.). Ultimately,
as Fischer above suggested, the general development was the same.
12 the OE ja-class which contained transitive verbs was no longer distinct from the intransitive oclass of weak verbs (Mustanoja 1960: 429)
13 I have not investigated external factors such as French influence in detail, but is is clear that
reflexive verbs borrowed from French (e.g. repent, remember, advise) were used with the simple
9
10
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
8
9.3.01
The term ‘reflexive verb’ is normally reserved for cases where the pronoun expressing
coreferentiality with the subject is ‘necessary’, i.e. obligatorily satisfies an argument
position of the verb. From today’s point of view, the many instances in OE of a socalled ‘pleonastic’ adjunct reflexive, mostly in dative case, seem messy because they
do not fit the standard definition of a reflexive construction as being a subcase of
transitivity. Kemmer (1993:39) points out that all current syntactic theories which make
use of the argument/adjunct-distinction run into problems because „valency is a
semantically-based phenomenon that only roughly correlates with syntax.“ Her analysis
of middle and reflexive constructions recognizes „the grammatical continuity between
participants integrally involved in the thematic structure of events, and those which are
more peripheral to the event“, thus replacing obligatorinesss (of arguments) with a
scalar notion: rather than being +/-obligatory, NP participants are more or less central
to the event denoted by the verb and therefore more or less prototypical. Compare:
(10) All to son, my brother, I fere me, for yow ‘even too soon for you, my brother, I’m afraid’
(Mankind 162, quoted after Peitsara 1997: 278)
(11) Horn him зede wel riзte.
He tok aþulf bi honde,
And vp he зede to londe.
‘Horn went indeed. He took Aþulf by the hand and went ashore’ (King Horn 69, quoted after
Peitsara 1997:319)
For a speaker of German, the translation of (10) above represents an instance of a
perfectly regular reflexively used verb; the example only sounds odd for a speaker of
Modern English, where ‘fear’ cannot be construed with a coreferential pronoun. (11) 14
would be unusual in German as well, although German still retains many verbs of
bodily motion used with a reflexive (sich setzen/drehen/wenden/hinlegen) that were
used similarly in Old and Middle English and have then become intransitive. With the
loss of case distinctions after OE and the dative-accusative syncretism in the pronoun
paradigm, the classification of coreferential pronouns into ‘necessary’ (accusative) and
‘unnecessary/pleonastic’ (dative) becomes useless, anyway.
Rather than discarding cases like (10) and (11) from consideration, the notion of
reflexivity should be enhanced to cover them as well, otherwise we run the risk of
prematurely narrowing the database. The common ground of ‘true’ and ‘pleonastic’
reflexivity is semantic; reflexivity as a semantic rather than a syntactic category
comprises both. Kemmer’s (1995) semantic scale ranging from the intransitive with
reflexive in English as well and did not prompt a „drift towards the intransitive form“, on the
contrary: von Seefranz-Montag (1983) has shown that French and English underwent radically
different developments: French developed reflexives from impersonals while English went the
other way round.
14 Note that the verb is used once with and once without a coreferential pronoun; the variation
seems to be due to metrical demands.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
9
9.3.01
only one participant involved to the transitive can serve to structure the semantic space
covered by the different reflexive constructions in ME:
Intransitive
middle
Reflexive
Transitive
one semantic role
two semantic
two semantic
roles, coreferent
roles, disjoint
„Middle voice event types comprise cases of intrinsic co-reference (i.e. co-reference
that is expected or predictable; thus for example, body actions such as ‘wash’,
‘dress’, ‘sit down’, etc. are semantically middle, not reflexive). With the middle, there
is only one semantic role inhering in the event, in contrast to the reflexive in which
two separate semantic roles are invoked, but whose participants happen not to be
referentially distinct. The middle, therefore, is characterised by a lower degree of
conceptual distinctness of participants than the reflexive; it approaches the
intransitive prototype in that respect“ (Kemmer 1995:65)
Both (10) and (11) would then be middle constructions, syntactically realized with a
coreferential non-argument reflexive. This pattern, although occasionally found in later
stages of the language,15 disappears, and it remains to be shown how the loss of this
possibility to code the middle is related to the rise of the new compound reflexive, a
point I will take up below.
Figure (1)16 depicts the distribution of the older reflexive strategy using the plain
personal pronoun compared to the newer one of using personal pronoun+SELF:
simple versus compound reflexive in Middle English (in %)
100
90
80
70
%
60
50
simple reflexive
40
compound reflexive
30
20
10
0
ME 1
ME 2
ME 3
ME 4
EModE1
EModE2
EMode 3
The diagram clearly shows the rise of the new compound reflexive at the expense of
the simple strategy. Puzzlingly, both appear with almost equal frequency in the first ME
period (1150-1250). From ME 2 onwards, pronoun+SELF gradually gain ground. The
turning point for the shift in dominance of the two strategies takes place between ME 4
(1420-1500) and EModE 1 (1500-1570), with an almost complete reversal of roles: the
15
16
e.g. when Ophelia is told to get thee to a nunnery!
Adapted from data in Peitsara (1997).
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
10
9.3.01
reflexive use of the compound form increases from below one third to two-third in the
next period and continues to establish its dominance in the following two periods, so
that in EModE 3 (1640-1710), the compound form is the rule. A more detailed
investigation of the texts from the transition period reveals that the rise of the
compound reflexive can be dated to the end of the 15th century:
„the SELF-strategy begins to gain ground in the 1480s (...) The increase in the use of
the SELF-strategy during the last two decades of the 15th century appears mainly in one
text type, i.e., sermons,17 which tend to favour this strategy in all periods. (...) The final
breakthrough of the SELF-strategy does not take place until the 16th century, and even
here its predominance is not marked until after the middle of the century.“ (288)
Keenan (1996) comes to a similar conclusion on the basis of his own corpus:
Percentage of local bindings expressed by -self forms (Keenan 1996:32)
99%
100%
77%
80%
87%
60%
40%
19%
24%
850-1066
1067-1200
20%
16%
20%
17%
1200-1303
1303-1405
1405-1495
0%
1495-1605
1605-1700
1700-1800
First of all, the figures clearly indicate that there was no sudden reanalysis of
pronoun+SELF such that the new compound reflexive became the only option of
marking reflexivity from one generation to the next.18 If we follow Keenan, then the
distribution of the compound reflexive remained more or less the same from Old
English onwards throughout Middle English, with the picture changing from Early
Modern English onwards. Peitsara on the basis of a much more refined corpus gives a
more specific date for pronoun+SELF becoming the dominant reflexive marker, but
notes that the emergence of the new form did not proceed uniformly, but was related
to the text type and the semantics of individual verbs. These last points suggest that
we are dealing here not with a syntactic change independent of context, but rather with
a process where the original meaning of the intensifier SELF was decisive in
determining the grammaticalization path.
17Other
text types which make frequent use of the complex reflexive are biblical texts and
statutes, which would be fully compatible with the semantic analysis given for SELF.
18 This is generally the main methodological problem for generativists working in historical syntax
and whose principal explanation for syntactic change is always a resetting of parameters which
are then passed on to the next generation of speakers; as far as I know, there is no syntactic
change that can be shown to have taken place within one generation.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
11
9.3.01
3.3.
Intensification
3.3.1. Attributive SELF
Old English, as already mentioned, did not differentiate between adnominal/adverbal
and adjectival/attritbutive intensifiers: SELF could be used as a possessive intensifier
synonymous with agen ‘own’. Agen, however, was restricted to possessive use; it is
related to the OE verb agan ‘to own, possess’, and developed in early ME to awe, owe,
to later become own.
The attributive use of SELF, modifying a head N, is hardly as frequent as its use as an
adjunct, but interesting insofar as it is in this position that independent SELF survives
into Middle English and beyond:
(12) A.Kn 2860 That in that selve grove, swoote and greene,
(13) B.ML 115 „Thy selve neighebor wol thee despise.“
(14) Bo 5. p3. 67/94 And herto I adde yit this thing: that ryght as whanne that I woot that a thing
is, it behoveth by necessite that thilke selve thing be; and eek whan I have knowen that any
thing schal betyden; so byhovith it by necessite that thilke same thing betide;
(15) 1539 (Cromwell in Merriman Life&Lett. (1902) II.175) As he knoweth right well, who at his
being here sawe her SELF visage (OED 905).
(16) 1632 (Lithgow Trav. IV. 158) They Gormandize at their selfe pleasures. (ibid.)
The relevant meaning of SELF here may be paraphrased as ‘uniqueness, singularity,
inalienableness’. If SELF is possessive, it expresses that something uniquely belongs to
someone. In a construction with a demonstrative or determiner, SELF intensifies the N
singled out by the determiner, intensifying the singularity, uniqueness of the entity
denoted by the N19.
König &Siemund (1998) note that intensifiers „always enforce a referential
interpretation of the NP with which they are in association“, which is also true of
attributive SELF.
3.3.2. Adnominal SELF
SELF
typically interacts with the most unique referent conceivable in the socio-historical
context of the Old and Middle English period, namely God. The expression swa swa
god sylf cwæð or swa swa Drihten sylf cwæð ‘as God himself said’ is practically
formulaic in the works of Ælfric.
SELF
typically intensifies a noun denoting a person of high standing, either because
they are unique and their position unquestionably given, such as God, the devil, or
19
cf. German derselbe, der Selbige, expressions which suggest that their main function is
reference tracking rather than intensification.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
12
9.3.01
because they occupy a high rank on the social scale, such as kings, bishops, apostles
etc.20 If the focus of SELF is a noun, it almost invariably refers to God or Christ, as has
been noted by all studies on the subject (e.g. Visser 1963, Ogura 1989). If the focus is
a pronoun, the link between the intensifier and its focus is less obviously determined by
the semantics of SELF as described above:21 to use Siemund’s (1997) terminology, a
pronominal focus does not have to be an organisational centre, but can be a situational
centre arising from the discourse context:
SELF was already used in OE after reflexive pronouns, e.g.
(17)(Or 128.6) Darius ...wolde hiene selfne forspillan ‚Darius ...wanted to destroy himself‘
The addition of SELF removes the ambiguity between the coreferent and the disjoint
interpretation of the pronoun in argument position. In (17) above, a typically otherdirected activity is carried out by the person denoted by the focus on himself.
Adnominal SELF structures a set of entities into a centre and a set of alternatives that
are peripheral to it: destroying is normally done by an agent to something or somebody
else, the agent thus forming the centre and the set of possible patients the periphery.
Without SELF, the more likely interpretation of the sentence would be that subject NP
and pronoun are disjoint. By intensifying the pronoun, SELF signals that the referent
designated by the pronoun is central, thereby reversing the expected agent-acts-uponpatient structure. Both subject NP and pronoun are marked as centre to the exclusion
of possible alternative values which facilitates the coreferent reading. Both in Old and
Middle English, SELF typically intensifies the simple reflexive of verbs denoting an
acitivity that is prototypically directed at somebody else, such as kill, destroy, hang,
murder, drown, and other unpleasant activities one does not normally do to oneself,
e.g.
(18)(LayA 5839) he makede him-seluen muchel clond ‚he made for himself much disgrace‘
Peitsara (1997:298) stresses that throughout Middle English, “verbs denoting
destructive and similarly negatively coloured actions that are usually not directed at
one’s self” as well as “verbs denoting other actions presented as undesirable or
condemnable” almost invariably take pronoun+SELF when used reflexively.
3.4.
The emergence of pronoun + self
Most if not all studies so far have concentrated on the development of the compond or
‘specially marked’ reflexive (van Gelderen 1998) and its subsequent syntactization.
20 SELF
is more frequent in religious than in secular texts, which could be due either to a) a close
adherence to the Latin original, from which most ecclesiastical texts were translated, and which required
the translation of ‘ipse’, or b) to the main protagonists of Christian texts, namely God, the devil, saints.
21
Keenan (1996) found that if the focus of SELF is a pronoun, the antecedent of the pronoun
refers to a remarkable being only in about 50% of all cases.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
13
9.3.01
The leading research question has been how a pronoun and SELF got together to
become the reflexive marker.
König & Siemund were the first to place that question within a wider typological
framework
of
intensification
across
languages.
In
their
analysis
of
the
grammaticalization of the English reflexive marker x-SELF, they stress that the relevant
question to be answered is that about the meaning of the two free morphemes which
finally ended up as a compound form: what was the semantic content which made
them capable of being interpreted reflexively? To put it differently, what kind of
implicatures did SELF give rise to that were compatible with the notion of reflexivity, and
in which contexts?
For the emergence of the compound intensifier, the other side of the coin, a similar
question has to be asked: why did a pronoun attach itself to SELF to form the new
intensifier - what were the relevant meanings of pronouns which may combine with a
free morpheme to signal intensification? Why was free SELF lost? The example of
Afrikaans shows that the development of a compound reflexive does not necessarily
entail the loss of the older simple form (K&S 1998:13)
From the very beginning up till well after the Middle English period, the semantics of
SELF
is sufficiently transparent to be the decisive factor in its use as compound
reflexive or intensifier. The continuous occurrence of simple and compound reflexive
forms, where the preference for SELF-forms is determined by the properties of the
referent denoted by the subject, is a strong argument against positing that SELF simply
underwent reanalyis around 1250 (van Gelderen 1998) and moved from a lexical into a
functional category, its distribution then only governed by syntactic requirements.
The common denominator of the attributive and the intensifying use of SELF seems to
be the signalling of uniqueness, understood as a property of referents on the one hand
and as a property arising from discourse on the other. SELF was an independent lexical
item that carried stress in alliteration in OE. At the end of the ME period, it had retained
so much of its semantic substance22 that it was not fully grammaticalized. However,
semantic bleaching and loss of positional variability had already affected SELF to
different degrees, according to its context of use: attributive SELF had kept its original
meaning and was still independent, but was eventually replaced by same and very.
Adnominal SELF adjoined to a nominal focus occurs only rarely after OE23; and it is only
in early texts where the pronoun is nominative rather than oblique when SELF is
adjoined to a pronominal focus, compare:
(19)(LayA 9214) 7 he seolf draf him forð
(20)(LayB 9214) and him-seolf drof heom forþ ‚And he himself drove him/them forth‘
22
23
for the term see Bybee et al. (1994.)
There are for example, only 7 instances of SELF intensifying a noun in the entire Layamon
text, the noun invariably denoting God.
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
14
9.3.01
Penning (1875:22) was probably the first to suggest that the modern compound form
x-SELF originated in the fusion of a ‘useless’ pleonastic dative him with SELF due to
their frequent occurrence adjacent to each other, a theory taken up by Mitchell: he
simply states that the modern compound form x-SELF developed out of dative him +
SELFNOM,
„the dative of a personal pronoun used more or less reflexively“...This
combination of dative of personal pronoun+nominative of SELF is, of course, the source
of some uses of MnE ‘himself’, ‘herself’, and ‘themselves’.(1985:196)“
It is tempting to assume, as Mitchell does, that examples like the following represent
early instances of the new compound form:
(21)ChronE (Plummer) (1087.57) ...& he sende of his mannan to þisum lande. & wolde
cuman himsylf æfter. ‘he sent [some] of his men to this land and wanted [to] come
himself afterwards’
Keenan (1996) adheres to a more refined version of this view; the source meaning of
the pleonastic dative pronoun would be ‚heightened involvement‘, while the presence
of SELF indicates ‚contrast‘, both referring to the subject of the predicate; so
semantically, the two forms would go together naturally. The loss of case distinctions
then created ambiguity as to which NP SELF was actually associated. To resolve this
ambiguity, the pleonastic pronoun as a pronominal copy of the NP was introduced
adjacent to SELF, and both elements then underwent the processes typical for
grammaticalization, namely phonological reduction and fusion.
The undisputable fact that pleonastic him and nominative SELF happen to occur quite
frequently next to each other is in itself not explanatory; frequency of occurrence is just
one
aspect
which
has
to
be
considered
when
looking
at
processes
of
grammaticalization, and it does not tell us anything about the meaning of the source
items in question.
K&S (1998) point out that the semantics of intensifiers offers one argument against
Mitchell’s analysis24: since intensifiers typically interact with referential NPs that are
definite descriptions, it is highly unlikely to assume that intensifying SELF was adjoined
to a non-referential pleonastic pronoun. Further, neither obligatory reflexives nor
reflexives marking derived intransitivity can be intensified, as illustrated with examples
from German25:
(22)Marianne schämt sich (*selbst). ‘Marianne is ashamed.’
(23)Paul wundert sich (*selbst), dass...’Paul wonders why...’
(24)Max dreht sich (*selbst). ‘Max is spinning around.’
(25)Dieses Hemd wäscht sich (*selbst) gut ‘This shirt washes well.’
If OE pleonastic him was an obligatory reflexive marker with some verbs and/or a
marker of derived transitivity, and given that precisely these reflexives block
24
shared by Penning 1875
25
Examples from project handout
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
15
9.3.01
intensification in present day languages, then Mitchell’s assumption that what is
impossible synchronically was possible diachronically becomes unlikely.
In (21) above, nominative SELF should be analysed as referring back to the antecedent
of the main clause, intensifying an unexpressed subject pronoun and therefore serving
the discourse function of reference tracking by singling out a unique antecedent. The
spelling of pleonastic him and SELF as one word is not a reliable clue for assuming that
the two forms were already fused at that stage; variation of spelling in the manuscripts
continues for centuries, and editors of texts often use one form when it resembles the
Modern English usage and two when it doesn’t, regardless of the original.
3.5.
The realignment of the middle domain
The term ‘transitive’ is usually used in relation with the verb valence and applied to
verbs selecting a direct object denoting the patient, the entity directly affected by the
action described by the verb. This most basic understanding of transitivity was refined
by Hopper & Thompson (1980). Since then, it has become quite common to regard
transitivity not as an absolute but as a scalar property of sentences as a whole,
allowing for higher or lower degrees of transitivity. They have broken down the
category of transitivity into several components or features which clausal elements
may possess to a varying degree. These parameters of transitivity taken together
make up a scale on which clauses may be ranked (ibid 252):
In this framework, reflexives are intermediate in transitivity between one- and twoparticipant clauses. Compared with one-participant clauses, they rate higher in
transitivity, in comparison to two-argument clauses, they are low on the transitivity
scale. Just as the middle and the reflexive are functionally and diachronically related26,
so are reflexive and impersonal constructions. According to Seefranz-Montag
(1983:77), the common denominator of both constructions is the ‘non-agentivity’ of the
object complement. In the reflexive construction, this lack of agent status is
syntactically marked as coreference of subject and object.
The grammaticalization of x-SELF is just one part of a larger process that spans the
whole of the Middle English period and concerns reflexivity, the loss of the middle
construction with a pleonastic pronoun and the loss of the impersonal. The last two
constructions rank lower on the transitivity scale than a reflexive, they can be used to
code less prototypical participants in a situation, to use Kemmer’s terminology. Modern
English has lost the abilty to express semantic differences between participants that
are more or less involved in the situation depicted by the verb; the only options left are
the reflexive, the intransitive and the passive construction; while German still retains
cf. Hermodsson, Velten (1931:327, quoted after von Seefranz-Montag 77): „to a certain extent, the
reflexive serves in the modern Germanic and Romance languages to express the middle voice as
distinguished from the passive proper.“ For the development of the passive, see Haspelmath (1990)
26
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
16
9.3.01
traces of the impersonal (as in mir graut vor dir) as well as the middle (as in Ich kaufe
mir einen Hut).
Different groups of verbs that took a pleonastic pronoun underwent different
developments in ME: verbs expressing motion (come, fare, flee, gan/go, haste, hide,
wend etc.) become intransitive by EModE (1640-1710). Verbs indicating a change in
body posture (bow, (a)stretch, prostrate, retire etc.) take the compound reflexive
(Peitsara 1997:315), verbs expressing emotion ((a)dread, awonder, fear, grieve,
repent) become intransitive. The so-called verbs of grooming (attire, bathe, cleanse,
clothe, dress warm, wash etc.) keep the middle marker throughout ME up to EModE2
(up to 1640); later, they are expressed either by intransitive constructions or by the
passive (ibid 335), e.g.:
(26)Go, and be thou waisschun in the watir of Siloe (Wycliffe and Purvey, The New
Testament John 9 11)
(27)Goo wesshe the in the pole of Syloe (Tyndale, The New Testament John 9 11)
(28)Goe, wash in the poole of Siloam (The Authorized Version John 9 11)
4.
Outlook: problems and perspectives
I have focused on one peculiarity of English, namely the polysemy of x-SELF, and have
tried to trace the origin of the modern situation. It turns out that the development of the
compound reflexive can be linked to a more general development, the realignment of
the transitivity continuum. The loss of the pleonastic reflexive construction and the
impersonal in English may be interpreted as the loss of a morphosyntactic possibility to
express lower transitivity: English sentence structure became less sensitive to the
transitivity continuum.
I have not said much about the loss of SELF as a free form, nor did I touch upon the
many questions related to the emergence of the intensifier x-SELF: why, for example,
could x-SELF be used as subject in ME, an option no longer possible? It is also not
quite clear yet whether the adverbal uses of x-SELF were present at all stages of
English and what role they played for the grammaticalization of SELF. These questions
can only be answered with detailed investigations of individual texts and their contexts.
Admittedly, typologists tend to be relatively careless when it comes to quoting
examples from older stages of the language, they are interested in the patterns
emerging from the text, not in the text itself. (Diachronic) typology necessarily involves
abstracting away from individual texts to arrive at generalizations over the data and
achieve explanatory value.
I think this is exactly where methodologically, typology as well as historical syntax and
‘philology’ don’t go together: both functionalists and formalists will happily ignore recent
research into the language of manuscripts in order to save their point, while the
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
17
9.3.01
philologist does not care much for syntax, let alone for more than one syntactic theory
or model in general27.
It is not feasible to reinvent the wheel and look at every single text in its manuscript
form; we would run the risk of losing track of the larger picture and come up with
nothing more than a collection of oddities, of confusing taxonomy of forms with
explanation. Panchronic typological evidence serves to structure the multiplicity of
language phenomena and to set up research priorities, and it would be unfair to
suggest that only the kind of data that fits the theory is taken note of; the
interdependence of theory and data being a truism for all science.
There is certainly a place for theory-driven accounts or sweeping generalizations of the
kind that invite frowns from more empirically oriented researchers as long as they
make for interesting hypotheses that can be tested independently. Examing data in the
light of such generalizations does not provoke a vicious circle, as Bybee et al. (1994)
have shown
“Everything that happens to the meaning of a gram happens because of the context
in which it is used. It is the use of language in context that shapes the meaning of
grammatical morphemes. Thus a true understanding of the mechanisms of change
that create grammatical meaning must proceed from analyses of the use of grams
as these changes are taking place.” (1994:297)
5.
References
Baker, Carl L (1995), „Contrast, discourse prominence, and intensification, with special
reference to locally free Reflexives in British English“. Language 71(2),: 63-101.
Bybee, Joan, Revere Perkins & William Pagliuca (1994), The Evolution of Grammar. Tense,
Aspect and Modality in the Languages of the World. Chicago: The University of Chicago Press.
Denison, David (1993), English Historical Syntax: Verbal Constructions. London: Longman
Denison, David (1998), „Syntax“. In: The Cambridge History of the English language Vol. IV:
1776-1997. Cambridge: CUP.
Fischer, Olga (1992), „Syntax“. In: The Cambridge History of the English Language vol. II: 10661476. Cambridge: CUP 207-408.
Haspelmath, Martin (1990), „The Grammaticization of Passives“. Studies in Language 14:25-70.
Hermodsson, Lars (1952), Reflexive und intransitive Verba im älteren Westgermanischen. Diss.
Uppsala.
Hopper, Paul & Sandra A. Thompson (1980), "Transitivity in grammar and discourse". Language
56: 251-299.
Howe, Stephen (1996), The personal pronouns in the Germanic languages. A study of personal
pronoun morphology and change in the Germanic languages from the first records to the present
day. Berlin: de Gruyter.
Keenan, Edward L. (1994/96), „Creating Anaphors: An Historical Study of the English Reflexive
Pronouns.“ Ms.
Kemmer, Suzanne (1993), The Middle Voice. Amsterdam: Benjamins.
My first point can be illustrated by van Gelderen’s (1999) neglect of the fact that the two Layamon
manuscripts have been shown not to be as far away from each other in time as previously assumed, the
second by the lack of space devoted to syntax in traditional OE/ME textbooks; Bruce Mitchell’s Old
English Syntax being a model of confusion rather than of grammar.
27
Reflexivity and Intensification in Middle English
Claudia Lange SEM III,
18
9.3.01
Kemmer, Suzanne (1995), “Emphatic and Reflexive –Self: Expectations, Viewpoint, and
Subjectivity”. In: Dieter Stein, Susan Wright & Edward Finegan (eds.), Subjectivity and
Subjectivisation: Linguistic Perspectives. CUP, 55-82.
König, Ekkehard & J. van der Auwera (1994), The Germanic Languages. Routledge.
König, Ekkehard & Peter Siemund (1996), "Emphatische Reflexiva und Fokusstruktur“. In: I.
Rosengren (ed.), Sprache und Pragmatik. Lund: Lunds Universitet.
König, Ekkehard & Peter Siemund (1996), „Selbst-Reflektionen“. In: Gisa. Harras (ed.), Wenn
die Semantik arbeitet. Festschrift für Klaus Baumgärtner. Tübingen: Niemeyer.
König, Ekkehard & Peter Siemund (1997), „On the Development of Reflexive Pronouns in
English: A Case Study in Grammaticalization.“ In: Uwe Böker & Hans Sauer (eds.), Anglistentag
1996, Dresden, Proceedings. Trier: Wissenschaftlicher Verlag, 95-108.
König, Ekkehard & Peter Siemund (1998), „The Development of Complex Reflexive and
Intensifiers in English.“ ZAS Working Papers, Berlin
König, Ekkehard & Peter Siemund (1999), „Intensifiers and reflexives: A typological perspective.“
In: Zygmunt Frajzyngier & Traci S. Curl (eds.) (1999), Reflexives: Forms and Functions.
Amsterdam: Benjamins. 41-74.
König, Ekkehard (1991), The Meaning of Focus Particles: A Comparative Perspective. London:
Routledge.
Mitchell, Bruce, (1979), „Old English self: four syntactical Notes.“ Neuphilologische Mitteilungen
Mitchell, Bruce (1985), Old English Syntax. 2 vols. Oxford: Clarendon Press.
Mustanoja, Tauno F. (1960), A Middle English Syntax. Helsinki: Société Néophilologique.
Ogura, Michiko (1989), Verbs with the Reflexive Pronoun and Constructions with Self in Old and
Early Modern English. Brewer.
Peitsara, Kirsti (1997), „The development of reflexive strategies in English“. In: Rissanen, Matti,
Merja Kyto & Kirsi Heikkonen (eds.), Grammaticalization at Work. Studies of Long-term
Developments in English. Berlin: Mouton de Gruyter p. 277-370.
Reinhart, Tanja & Eric Reuland (1993), „Reflexivity.“ Linguistic Inquiry 24(4): 657-720.
Siemund, Peter (1997), Intensifiers. A Comparison of English and German. Berlin: Inaugural
Dissertation.
Traugott, Elizabeth Closs (1992), „Syntax“. In: The Cambridge History of the English Language
vol. I: The Beginnings to 1066. Cambridge University Press. 168-289
van Gelderen, Elly (1998), A History of Reflexive Pronouns: Where Pro-Drop and Agreement
meet Case. Ms.
van Gelderen, Elly (1999), „Bound pronouns and non-local anaphors: The case of Earlier
English”. In: Zygmunt Frajzyngier & Traci S. Curl (eds.) (1999), Reflexives: Forms and
Functions. Amsterdam: Benjamins. 187-225.
Venezky et al. (1980), A Microfiche Concordance to Old English. Toronto.
Visser, F.T. (1963/73), An Historical Syntax of the English Language. Leiden.
von Seefranz-Montag, Ariane (1983), Syntaktische Funktionen und Wortstellungsveränderung.
Die Entwicklung ‘subjektloser’ Konstruktionen in einigen Sprachen. München: Wilhelm Fink.