Content uploaded by Sanchez-Stockhammer Christina
Author content
All content in this area was uploaded by Sanchez-Stockhammer Christina on Dec 20, 2017
Content may be subject to copyright.
Chapter 9
Hybridization in Language
Christina Sanchez-Stockhammer
Abstract Hybridization is a phenomenon that can be observed in many cultural
domains – not least in language. After a consideration of the term’s origins,
hybridization is defined as a process whereby separate and disparate entities or
processes generate another entity or process (the hybrid), which shares certain
features with each of its sources but which is not purely compositional. The paper
then considers possible instances of hybridity – the basis for hybridization – on
different levels of language, such as speech sounds, words and texts. It posits that
hybridization is possible on all levels of language, from the most basic to the most
abstract, but with regard to different aspects, namely formal, semantic, functional,
etymological and communicative hybridity. The frequently used metaphor of
language as an organism may explain the closeness of linguistic hybridity to the
original biological concept – though particular features of the system language,
such as the distinction between the levels of langue and parole (cf. de Saussure
1916/2005, 30–31), give hybridization in language a special character.
9.1 Is Hybridisation Hybrid?
At first, the question above seems to make little or no sense – at least not if
it is encoded orthographically the way it is. However, if we follow the conventions
which are usually adhered to in linguistics and italicize hybridization, the question
starts to make sense indeed, since it does not refer to the process of hybridization
anymore, but to the word itself. From being a question about the world it has
come to be a question about language, a metalinguistic question – whose answer
will need to be postponed because it first requires a definition of what is understood
by a hybrid.
Originally, the term hybridization comes from the domain of biology. Since the
middle of the nineteenth century, it has designated the production of hybrids, i.e. of
P.W. Stockhammer (ed.), Conceptualizing Cultural Hybridization,
Transcultural Research – Heidelberg Studies on Asia and Europe in a Global Context,
DOI 10.1007/978-3-642-21846-0_9, #Springer-Verlag Berlin Heidelberg 2012
133
animals or plants which are “the offspring of individuals of different kinds (usually,
different species)” (SOED, i.e. Shorter Oxford English Dictionary, s.v. hybrid).
A typical example would be the mule, which is neither a horse like its mother nor a
donkey like its father (SOED s.v. mule).
According to Ackermann (2004, 141), the term hybrid (which was coined in the
early seventeenth century; cf. SOED s.v. hybrid) was rarely used before the
nineteenth century – during which it was loaded with very negative connotations,
based on the assumption that this process was detrimental to a supposed original
state of purity and authenticity, and mainly with reference to the biological context
of mixed breeding (Fludernik 2005, 227). One reason for this negative view
may lie in the fact that hybrid organisms are often (though not always) infertile
(Weißk€
oppel 2005, 317); another in the assumption, propagated by Gobineau and
other racist theorists, that whenever there was interbreeding, the inferior genes
would prevail over the superior genes (cf. Nederveen Pieterse 1999, 177). In
addition, hybridization clearly opposed the nineteenth-century and structuralist
obsession with taxonomies (Bucakli and Reuter 2004, 174) because hybridity, by
its very existence, blurs the distinctions among categories (Nederveen Pieterse
1995, 56). Since the development of Mendelian genetics in the 1860s (Blech
et al. 2005, 140) and particularly since the twentieth century, however, a
re-evaluation has taken place which regards cross-breeding and polygenetic inheri-
tance as valuable enrichments of the gene pool (Nederveen Pieterse 1995, 55;
Ackermann 2004, 141). The concept of hybridity theory was then applied to
language by the Russian literary and linguistic scholar Mikhail Bakhtin and later
extended to refer to many different domains – first in linguistics, and subsequently it
was turned into “a positive label of multiculturalist racial intermixing, syncretism,
and transnationality” (cf. Fludernik et al. 2005, 228) by post-colonial critics.
Weißk€
oppel (2005, 334) differentiates between the terms hybridization,syncretism
and creolization. According to her, the defining feature of syncretism is the idea of
blending, whereas creolisation is dominated by the observation of coexistence and
combinations of different weighting. In Weißk€
oppel’s terminology, hybridity
emphasizes the status quo in the contact situation, the unfinished and the parallel.
However, not everyone would subscribe to this distinction. According to Nederveen
Pieterse (1999, 178 and 2001, 237), the term hybridization is more common today
in humanities as a general term than its possible alternatives bastardization,mixing,
blending,melding,merging,syncretization,creolization,me
´tissage,mestizaje,
cross-over etc. Even so, as the idea of combining disparate, seemingly pure entities
to form a new, different one still seems to be widely regarded as fairly negative – for
instance, bastard is used as an insult in different languages (e.g. English and
German), and the corresponding verb bastardize means ‘cause to deteriorate’ or
‘stigmatize as illegitimate’ (SOED s.v. bastardize), – it may not be long before
hybrid acquires the same connotations, so that a new, more fashionable term will
have to be coined to designate this increasingly important process in our globalizing
world once again in a neutral way.
134 C. Sanchez-Stockhammer
9.2 Hybridity and Hybridization in Language
Before the concept of hybridization can be applied to the domain of language, we
need a definition of hybridization: it will be understood here to be a process
whereby separate and disparate entities or processes generate another entity or
process (the hybrid), which shares certain features with each of its sources but
which is not purely compositional. By the introduction of non-compositionality, we
avoid treating entities as hybrids which are produced by an additive relationship,
e.g. in the sense that three is two plus one. Not only does this definition represent a
synthesis of the definitions of hybridization one encounters in linguistics but it is
general enough to be applicable to a wide range of phenomena as well. It builds
very much upon the original, biological concept of hybridization, just like most
linguistic definitions of the concept seem to do.
The next step in the consideration of hybridization in relation to language is
necessarily the contemplation of the levels of language which may be affected by
hybridization processes. The triangle in Fig. 9.1 provides an overview of these
levels, with the smallest elements at the bottom and the largest ones at the top.
1
By its very nature, the definition of hybridization relies to a very large extent on
what is understood by a hybrid. For this reason, we will now search for linguistic
communication
languages
text types
texts
sentences
clauses
phrases
idioms
collocations
words
morphemes
sounds
Fig. 9.1 The levels of language
1
It will be noted that though hierarchical, this overview of language is characterized by a certain
degree of overlap (e.g. idioms, collocations and phrases may be of varying size), and that
communication on the highest level is a process rather than a tangible constituent. This is due to
the fact that the higher we move up in the stratification of language – particularly beyond the
sentence level –, the more difficult it is to decide what the structure is supposed to be (cf. Bolinger
1975, 16).
9 Hybridization in Language 135
hybrids on all levels of language. The examples will be taken from English – but
they could also be taken from any other language.
9.2.1 The Level of Speech Sounds
Speech sounds represent the most basic level of language. Since they are by
definition the smallest units that language is composed of on the formal acoustic
level, it is hard to see how a concept like hybridity, which involves a combinatory
process, should be applicable at all. Although originally foreign speech sounds may
be included into a native system, this merely involves the addition of new items to a
pre-existing entity, so that no parenting relation such as that required for hybridi-
zation can be recognized.
2
However, as soon as we leave the level of individual
sounds in the strictest sense and consider phenomena of connected speech, we may
encounter hybridization.
Assimilation is the process whereby more or less neighbouring sounds become
more similar to each other in order to make articulation easier (Gut 2009, 35). Thus
when pronouncing the word unkind, speakers often realize the second sound as [ŋ]
instead of [n] because of the following [k]. The sound [ŋ] thus shares features of
both [n] and [k]: [k] is a velar plosive, [n] an alveolar nasal – and [ŋ] a velar nasal. In
other contexts, though, the sound [ŋ] must be regarded as a single, basic sound – e.g.
when it occurs as the last sound in sing. It would make little sense to speak of a
hybrid there. Otherwise, practically all speech sounds would need to be considered
hybrids on the basis of similarities of their articulatory features to other sounds.
Nonetheless, new phonemes may eventually emerge from continued assimilation:
thus the palato-alveolar English fricative /ŋ/ is the result of a combination of the
alveolar fricative /z/ and the palatal approximant /j/, which emerged in contexts
such fusion as (Scheler 1977, 79). This means that we must differentiate between
two ideas of hybridization, namely first a context-independent one, which refers to
the makeup of linguistic entities, and second a context-dependent one, which refers
to their use in particular contexts.
This initial example also raises some further questions. For instance, how
important is it that the sources should be recognizable in a hybrid in order to define
it as such? The velar nasal can obviously be traced back to its two “parents” in an
articulatory-feature analysis, but the relationship will presumably be unnoticeable
to an uninitiated listener. As a logical consequence, an entity for which it is
impossible to detect different sources by any means can never be called a hybrid
because a very important prerequisite is missing. However, the parenting relation-
ship need not be immediately recognizable, and any proof – be it visible to the
2
Note, however, that this does not preclude the existence of hybrid pronunciation systems
according to Wawra (this volume, for New Zealand English).
136 C. Sanchez-Stockhammer
naked eye or only after an analysis by experts using special instruments – will
suffice.
Careful observers will have noticed that the example above still poses another
problem. If the definition requires entities to jointly give rise to a new one, does that
involve any implications that the sources must remain untouched, or is it acceptable
if one of them ceases to exist and is transformed into the hybrid, as the /n/ in unkind,
whose place is filled by the newly-emerged /ŋ/, while the /k/ remains in its usual
place? One may argue in favour of such a view because language is a very special
system by its twofold nature, which consists of the langue (the system) and the
parole (the actual use of language) (cf. de Saussure 1916/2005, 30–31). Even if the
source is transformed in a process of hybridization in this particular case, it
therefore still exists on another level.
9.2.2 The Level of Words: Morphemes and Words
Hybridization being a process, the question to consider on the lexical level is
by which processes new words come into a language. Basically, there are three
possible alternatives: either new words are formed from scratch, or new words are
formed on the basis of previously existing material, or they are borrowed from other
languages. In the first case, hybridization is definitely not possible: if there are
no separate sources, we cannot speak of hybridization. Thus the English word bed
is reasonably short, cannot be subdivided any further and must have been the
result of a creative formation from scratch at some time in the very remote past.
By contrast, in the case of word-formation processes, basic linguistic units – the
so-called morphemes, which are the smallest units of a language with lexical or
grammatical meaning (cf. Lipka 2002, 3) – are combined into larger sequences,
namely words. Thus waterbed is a compound formed from the combination of
water and bed.
9.2.2.1 Meaning
At first sight, this does not seem to be any more than an additive relation: waterbed ¼
water + bed, but there is actually more to it. For instance, a determining relation
holds between the two constituents, with water describing bed more precisely.
While this could still be captured by language-immanent rules of the type “When
two nouns are juxtaposed, there may hold a relation in which the first acts as a
modification on the second”, what no generally applicable rule can grasp is the
precise relation between the two constituents. There are certainly a number of very
typical relations between constituents of compounds – e.g. ‘a kind of’, ‘a part of’
etc. (cf. e.g. Levi 1978, 76–77) –, but it is not possible to formulate a rule that would
encompass any reasons for a selection from a predetermined amount of relations. In
the case of waterbed, the constructed meaning ‘a bed made of water’ comes fairly
9 Hybridization in Language 137
close to the word’s usual meaning in the language, but is still not precise enough. In
other instances, the relations between the constituents are even more idiosyncratic.
For this reason, we are actually dealing with semantic hybrids here, as the rule-
supported addition of the parts is not enough to arrive at the lexicalised result of
the word-formational process (cf. Bauer 1983, 48–50). This is not only true
of compounding but also of other processes such as suffixation, e.g. in the case of
drawer. One of the meanings of to draw is ‘to pull’. The suffix -er is typically used
to convert verbs into nouns; in this particular case we can recognize the meaning ‘a
thing suitable for’ (cf. SOED s.v. -er
1
), so that a drawer is ‘a thing suitable for
pulling’. Nonetheless, it is still a certain step from here to the usual meaning of the
word, which involves the meaning component ‘part of a piece of furniture’. We can
therefore argue that whenever an idiomatic component is involved in the formation
of a new word, we are dealing with semantic hybridization.
But does that mean that all words consisting of several morphemes should be
defined as hybrids because there is always some kind of semantic change, i.e. a
change in meaning, when morphemes come together? No; whenever it is possible to
capture the new meaning by the application of rules, we should not speak of
hybrids. For instance, if we add the regular plural morpheme {S} to the word
bed, it attains a plural meaning: beds. If we add this morpheme to car, we get cars –
the same word, only in the plural. This is an entirely regular process. Of course one
may object that this is no word-formation but only the addition of grammatical
meaning by means of a grammatical morpheme, {S}. Still, this regularity also
occurs with certain lexical morphemes, e.g. with the prefix un-.
3
When attached to
an adjective, it negates its meaning in a fairly regular fashion, e.g. reasonable to
unreasonable. Yet even so, there may be certain changes: the Longman Dictionary
of Contemporary English gives more than ten meanings of the word clear, e.g.
‘easy to understand’, ‘impossible to doubt’, ‘sure about something’, ‘easy to see or
hear’, ‘clean and fresh’. Unclear, by contrast, is reduced to a single meaning
involving difficulty in understanding. For this reason, a person’s complexion can
be clear but not unclear (though there are a number of hits for this collocation on the
internet, too). Nonetheless, unclear is still compositional, if only with regard to one
of the lexical units of clear (Cruse 1986, 80), and thus not hybrid.
Though hybrid word formation may be based on the precise number of two
sources, it is of course possible that more than two morphemes are involved in the
creation of new words with a semantically hybrid meaning. However, their combi-
nation usually follows a particular order, so that the process can be considered as
a combination of binary steps, in which the so-called immediate constituents
(cf. Kastovsky 1982, 170) are combined: thus ice cream cone consists of ice
cream + cone, the first of which is a hybrid that can be analyzed into the hybrid
3
Gauger’s (1971, 133) verschiebend (‘shifting’) and variierend (‘varying’) durchsichtige W€orter
are word pairs such as French tendresse ‘tenderness’/tendre ‘tender’, which have the same meaning
except for the fact that they differ in their part of speech, and maisonnette ‘small house’ vs. maison
‘house’, where the complex word only represents a specific variation of the base.
138 C. Sanchez-Stockhammer
ice + cream. A reversal of this order (ice + cream cone) usually produces
nonsense due to the default rule of English word formation according to which
the first part of a compound determines the second part (cf. Lipka 2002, 96; though
there are of course some exceptions). A similar binary order can be observed in
affixations.
From a semantic point of view, polysemy and homonymy may also initially
seem good candidates for semantic hybrids. By polysemy we understand that one
lexical item has several, related meanings (cf. Lipka 2002, 92), e.g. lay ‘put down
into a flat position’ and ‘produce an egg’. Homonymous words, by contrast, have
the same form but very different meanings, e.g. ear is not only used to designate a
part of the body but also a head of corn (Lipka 2002, 154). However, a closer look at
these phenomena reveals that they have one thing in common which distinguishes
them from hybridity, namely the fact that the elements are not fused but merely
alternatives, which can still be kept apart and may be actualized in different
contexts. A hybrid, by contrast, is fed from different sources but clearly a single
entity. Whenever different meanings share an identical form in the widest sense, we
should therefore not speak of hybridity.
9.2.2.2 Form
This leads us directly to the formal side of words. In compounding, the two chains
of sounds/letters representing the formal side of the morphemes involved usually
remain untouched. Affixation will sometimes result in fairly regular formal
changes, e.g. the deletion of the verb-final <e>in drive in the noun driving, but
this should not be regarded as an instance of hybridization because of its rule-like
character. Some other word-formational processes, though, result in more unpre-
dictable shapes. Blends as non-morphematic word formations (Fandrych 2008,
109) are a particularly interesting case in point. Thus a brunch is neither breakfast
nor lunch; though it does involve food typically associated with either, it is usually
consumed at a time starting between the two and typically in other social contexts
(the corpus sentences in the LDOCE suggest that brunch is usually consumed
outside the home, and typically during holidays). Not only do blends combine
more or less random parts of their source lexemes’ form,
4
but their meaning also
combines semantic features of both sources, so that they can be considered iconic
in that their forms reflect their referents (Fandrych 2008, 113). This type of word-
formation is particularly convenient in the naming of hybrid plants or animals
(cf. Meredith 1948) because it can be considered hybrid itself.
4
There are certain rules that are applied when words with particular properties are combined with
each other (cf. Plag 2003, 122–126).
9 Hybridization in Language 139
9.2.2.3 Etymology
But let us come back to the issue of how new words enter a language. The third way
is represented by borrowing from other languages. The reasons for such processes
of borrowing are manifold (Scheler 1977, 86–88): thus words may be taken over at
the same time as a foreign concept, process or entity, e.g. English sauerkraut from
German. Furthermore, it may be that an existing loanword in a language opens the
way for related borrowings from the same word family, as is the case with the
Middle English French borrowing judge, which is complemented by the French
borrowing just in late Middle English (SOED s.v. judge,just; Scheler 1977, 87).
Sometimes the foreign origin is still obvious, as in British bureau (which is a
large desk, just like in French), and whose borrowed spelling <eau>strikes the
eye. Its quasi-synonym desk, by contrast, may be less easily recognizable as
ultimately going back to Latin discus (SOED s.v. desk) – which means that this
word has undergone massive changes in both its form and meaning. As both words
were borrowed as a whole, they cannot be considered hybrid in the sense discussed
before, though – in spite of the fact that the foreign word discus (or its medieval
Latin form desca; cf. SOED s.v. desk) was modified by generations of usage on the
British Isles before becoming modern desk: the Latin and English elements which
combined to yield this modern English word are of very different types, one of them
being an entity (the word) and the other one representing a process. Though it is
conceivable that two different processes may engender another, hybrid, process,
just as two entities can engender a hybrid entity, it is highly questionable whether
the interaction of an entity and a process should be regarded as resulting in hybridity –
otherwise almost everything would have to be considered hybrid in a certain way.
Similar problems arise when we consider the morphological changes arising
when foreign words are integrated into a language and become loan words through
this process (Scheler 1977, 89). For instance, verbs may be stripped of their inflec-
tions and/or get new ones from the target language system. While we have seen
before that the rule-based addition of grammatical morphemes to word stems
cannot be considered an instance of hybridization, the picture becomes some-
what different when the two sources come from distinct languages. Thus, unlike
French pre
´senter, English present no longer has an explicit infinitive ending –
however, it forms the past tense with -ed (presented), the continuous form with -ing
(presenting) etc. As these grammatical morphemes are of Germanic ancestry
(cf. SOED), presenting can therefore be considered hybrid from an etymological
point of view. However, this raises yet another question: the original French word
had an infinitive ending and the English one does not. Can present in its bare form
then also be considered a hybrid? As we have stated before, the stripping of the
suffixes is a rule-based process in contrast to the entity-like character of the word.
For the reasons outlined above, we should consequently not consider this a hybrid.
5
5
The only way to do this would be by postulating a morpheme that goes one step beyond the zero
morpheme (cf. Lipka 2002, 3), which we may call a negative morpheme {X}. This morpheme is
140 C. Sanchez-Stockhammer
Still, the interaction between different languages may result in hybridity when
new words are formed on the basis of previously existing material from different
etymological sources. This is a use of hybrid that we find very commonly – and also
very early – in the linguistic literature (e.g. in McKnight 1923; Wehrle 1935).
As these hybrids are formed by speakers who will usually belong more to one
linguistic community than to another, the results of such processes will presumably
be integrated into the stronger language.
The ultimate example for etymological hybridization in English is given by
McKnight (1923, 161): remacadamizing consists of the following constituents:
6
•re- Latin
•Mac Celtic
•Adam Hebrew
•-ize French <Latin <ultimately Greek
•-ing Germanic/native English.
McKnight’s example also demonstrates very clearly that from an etymological
perspective, linguistic hybrids may have more than two sources without any
problem. Etymologically hybrid word formations are attested to in different
languages and varieties (cf. e.g. Schach 1948 for Pennsylvania German and Irwin
2005 for Japanese) but seem to be particularly frequent in English (Wehrle 1935,
60–61) due to its external language history: in 1066, the French-speaking Normans
invaded the isles and had a strong impact on the language for several centuries
(Crystal 1995, 30–31). This paved the way for a massive influx of Romance words
into the language. Such integration is possible in different ways of word formation,
such as compounding (doubletalk <French double, Germanic talk), prefixation
(besiege <Germanic be-, French siege;dislike <Romance dis-, Germanic like)
and suffixation (priesthood <Latin priest, Germanic -hood; husbandry <Old
Norse husband, French -ry). Nonetheless, Wehrle (1935, 5) reports that there are
relatively few etymologically hybrid words in English before the late thirteenth
century. The first attested hybrid affixation is sotship – around 1050 –, which later
died out (Wehrle 1935, 17). Apart from formations in -ship, only very few other
Germanic suffixes, such as -hood,-ness or -ful, combined with French bases before
1300 (Wehrle 1935, 20). The first Germanic prefix to combine with a French base
was be-,inbespouse (Wehrle 1935, 20–21). Up until about 1300, hybrid affixations
consisted exclusively of French loan words and native English affixes (Wehrle
1935, 57–58) – a process that Wehrle regards as organic and subconscious.
similar to the zero morpheme in that it serves to complete inflectional paradigms and has no form
but a meaning. In the example above, it stands in place of foreign infinitive endings, thereby
causing their deletion – which explains its name.
6
Though this is a nonce word, cf. the SOED for the meaning of its constituent macadamize: “Make
or repair (a road) according to McAdam’s system, by laying down successive layers of broken
stone of nearly uniform size, each layer being consolidated by pressure before the next is laid.”
9 Hybridization in Language 141
Sch€
onfelder’s (1956, 65) explanation for the fact that we do not find the reverse
pattern (Germanic base and French affix) at the same time already is that affixes are
not borrowed directly from one language into another but that they enter a language
by the analysis of loan-words and analogical formations.
The so-called loan formations (Lehnbildungen) are a very special case of
etymological hybridity: though their form is native, their structure and meaning
come from another language (Scheler 1977, 92–95). For example, English folk song
was created in complete analogy to German Volkslied. Scheler (1977, 93) calls this
subcategory of loan formation Lehn€
ubersetzung (loan translation), in contrast to
the slightly less literal Lehn€
ubertragung (loan transfer), such as wishful thinking,
whose German source Wunschdenken has the literal translation ‘wish thinking’.
Lehnsch€opfung (loan creation), the third subcategory, is even freer, in that a word in
a foreign language stimulates the creation of a new word with an identical meaning
but a different form (e.g. German vollklimatisiert <English air-conditioned)
(Scheler 1977, 93–94). Consequently, the loan translations are etymologically
hybrid but in an unusual way, since the etymologically alien element may either
be semantic only or refer to the structure to a greater or lesser degree. The so-called
Mischlehnw€orter (mixed loan words; Scheler 1977, 90) cross even more boundaries:
while they borrow the meaning, structure and one of their components from a foreign
language, the other component is native (e.g. German Grapefrucht for grapefruit;
Frucht ¼fruit). The opposite trend can be observed in the Lehnbedeutung (loan
meaning): here, only a foreign word’s meaning is added to an already existing native
word, e.g. German f€
uttern ‘feed’ was extended to mean ‘insert data into a computer’
in analogy to English feed (Scheler 1977,95).
So far we have unquestioningly accepted the common assumption that words
may be etymological hybrids. However, this idea can be questioned by arguing that
if an English word-formation element combines with a French one, for example the
result is still an English word. The word may be hybrid in different ways (e.g. with
regard to its form, meaning, and the fact that it comes from two sources), but it does
not constitute an instance of a new language. However, we may counter that this
reverses the perspective: the combination of word afrom language Awith word b
from language Bhas to yield word c– but not necessarily in a new language, C.This
can be regarded in analogy to biological hybrids: a mother with feature A and a
father with feature B will yield a hybrid offspring – but still either male or female
(at least usually).
According to Nederveen Pieterse (1995, 55–56), “hybridity concerns the mix-
ture of phenomena which are held to be different, separate; hybridization then refers
to a cross-category process.” This engenders the question how distinct the sources
actually have to be, so that another way of questioning the English hybrids would
consist in enforcing stricter rules regarding the issue of proximate vs. original
language. Thus French and Latin are different languages, but ultimately French
goes back to Latin; similarly, English and French are both ultimately Indo-European.
We may therefore argue that in a very broad sense, English-French compounds are
not really etymologically hybrid if we wanted to restrict the term to combinations
transgressing larger language families.
142 C. Sanchez-Stockhammer
9.2.3 The Level of Fixed Constructions: Collocations and Idioms
Several levels lie between the level of words and that of sentences. Since these overlap
to a certain extent, we shall first consider the level consisting of more or less fixed
combinations of words, namely collocations and idioms. Collocations are words that
typically occur together, such as to take a shower (and not to *do a shower (cf.
Hausmann 2004, 309 for German examples). There are two approaches towards
collocation, one of them considering a significantly high statistical frequency of co-
occurrence (cf. Bartsch 2004, 91), and the other implying that a special relationship
obtains between two words (cf. Hausmann 2004, 320–321; Hausmann 1985,118).Be
that as it may, words that typically occur together – and make language sound “right”
to native speakers – very often contain some idiomatic component. This is even
stronger in the case of idioms, whose meaning cannot usually be guessed from the
meaning of their parts: e.g. to kick the bucket means ‘to die’ (Flavell and Flavell 1992,
117). In contrast to words, idioms and collocations are therefore never hybrid on the
formal level (because they represent combinations of whole words) – but idioms (and
to a certain degree collocations) are necessarily semantically hybrid.
In addition, the most suitable and most frequent collocation of a particular word
may have a very different etymological source from the word itself: table is Romance,
but the process of preparing a table for a meal is not expressed by the equally Romance
verbs prepare or decorate; instead, we say either lay the table or set the table (cf.
Hausmann 2004, 309 and SOED s.v. table), thereby generally combining the Romance
noun with a Germanic verb. Such collocations, of which there are presumably many
more, show how intricately the Germanic and Romance elements interact in English.
For this reason, we may even speak of a very weak form ofetymological hybridity here.
9.2.4 The Level of Syntax: Phrases, Clauses and Sentences
The same seems superficially to be true of phrases, as one can speak of a red car or
an orange car without any restrictions, even though the word car is Romance but
ultimately of Celtic origin, while red is a Germanic word and orange has entered
English from Sanskrit via Persian, Arab and Old French (cf. SOED). We can use
words with different etymologies freely within phrases, clauses and sentences.
Syntactic rules are thus entirely blind towards etymological origin. But should
this be considered a form of hybridity? No. Due to the mainly rule-based, quasi-
compositional nature of syntax, this level has little to offer in terms of
hybridization – as long as there is no overlap with collocations or idioms, which
may take different syntactic shapes, from the combination of premodifying
adjectives within a noun phrase to sentence-length proverbs such as “When in
Rome do as the Romans do”, to practically anything in between.
Still, this does not mean that functional hybridity – as an alternative to semantic
hybridity in the realm of grammatical structure – is impossible. For instance, the
9 Hybridization in Language 143
construction traditionally called a gerund combines both verbal and nominal
characteristics (Quirk et al. 1985, 1290–1292 and Herbst et al. 1991, 107–108):
the word building in They admired his quickly building the tower is like a verb in
that it is premodified by an adverb and followed by a direct object but like a noun in
that it is preceded by a possessive determiner. Other gerunds – such as writing in the
art of writing poetry – may display the additional nominal characteristic of being
the head of a prepositional phrase. As these two functions of the gerund are
inseparably linked with each other, we can consider it a functional hybrid.
There are even more such examples. Thus English grammar recognizes two verbs
dare (Quirk et al. 1985, 136–139): one main verb with full inflection and do-support
(He doesn’t dare to escape.) and one auxiliary verb with bare infinitive and no
do-support (He daren’t escape.). Still, Quirk et al. (1985, 138) point out that blends
between these two constructions occur and are widely acceptable, e.g. They do not
dare ask for more – which combines the do-support of the main verb use with the bare
infinitive typical of the auxiliary. Thus features of both categories are present in such
a construction, without it belonging to either of the two. Not only is this an instance of
functional hybridity but also of gradience (cf. Quirk et al. 1985, 90), i.e. the fact that
an item does not clearly belong to one category or another but is situated in between.
Instances of functional hybridity are thus always gradient – but in the special sense
that they are situated at a more or less equal distance to the two ends of the scale
(while gradients that are closer to one particular end would rather be regarded as
being atypical instances of that particular type). As grammatical categories are
always a matter of definition (cf. the different categorization of parts of speech in
different grammars), the determination of functional hybridity will be particularly
dependent on the grammatical model used as the basis for categorization.
9.2.5 The Level of Text: Texts, Text Types and Genres
If we move yet another step higher, etymological hybridity may also seem to be
observable to a certain degree in texts. However, one must question whether words
may be considered as directly generating hybrid texts – after all, phrases, clauses
and sentences function as intermediate steps, and the definition of hybridization
does not provide for the omission of the parent generation and the jump from the
grandparents to the grandchildren. In addition, we must not forget that the interme-
diate syntactic steps between words and texts are rule-based (unless we are dealing
with fixed expressions, which are usually idiomatic to a certain extent; cf. above)
and therefore not really of interest to hybridization, whereas it is not really possible
to ascribe a particular etymology to individual sentences.
However, we can observe a new kind of hybridization on the level of texts,
namely the emergence of new text types from the combination of existing ones.
Galster (2005, 227) uses the term hybrid genre to designate “works of art which
transgress genre boundaries by combining characteristic traits and elements of diverse
literary and non-literary genres”. This might result in the re-classification of more
144 C. Sanchez-Stockhammer
traditional text types as hybrid genres, but more usually in the description of new text
types in which elements of one genre appear in another, e.g. when a laudatory speech is
blended with an appeal (Jamieson and Campbell 1982, 147), or when talk show
elements are introduced in political interviews (Lauerbach 2004). Promotional news
represents a mixture of journalism and advertising, which is reflected in the blend
advertorial (Erjavec 2004, 554). Hybrid novels may break down the boundaries
between fiction, poetry and drama (Galster 2005, 227). Some genres have exerted
such a strong influence on the structure of particular novels that the emerging novel-
types were named after them, e.g. the confession, the diary etc. (Bakhtin 1981,321).
7
9.2.6 The Level of Individual Languages
In narrative studies, as opposed to linguistic usage, the term hybrid novel is used with
reference to “novels in which Western and post-colonial (native) writing traditions
creatively interact” (Fludernik et al. 2005, 227). This usage of the term hybrid thus
refers not so much to an intermixing of genre but rather to the merging of cultures. It is
one step further away from the linguistic use of the term, which relies more on the
original, biological definition. But this is no surprise, because language, the subject of
linguisticstudy, is often – particularly in the later nineteenth century, and modelled on
Darwin’s ideas – likened to an organism, which evolves, dies etc. (cf. Gut 2009,7).
This is still recognisable in the fact that linguists have for a long time been talking
about language families (cf. Bolinger 1975, 457), i.e. languages that are “genetically”
related to each other (Bußmann 1990 s.v. Sprachfamilie), and which may be depicted
in family trees, in analogy to human families (e.g. Bußmann 1990, 277). This may be
the reason why the idea of hybrid languages is one that is fairly frequently used, even
in the older linguistic literature. In this context, it is interesting to note that hybridity in
language differs from hybridity in biological organisms in that there seems to be no
restriction on the possible combination of languages, whereas biological organisms
need to be structurally relatively similar to produce offspring (Croft 2000: 209). Even
though English is certainly not the only hybrid language (cf. e.g. Wurm 1995), English
is often considered “the standard example of a hybrid language” (Roberts 1939,23).
8
7
While the former are either written texts – or at least may be written –, we can apply the concept
of hybridity to spoken texts as well: thus, Kamberelis’ (2001, 86) hybrid discourse practice
involves “teachers and children juxtaposing forms of talk, social interaction, and material practices
from many different social and cultural worlds to constitute interactional spaces that are
intertextually complex, interactionally dynamic, locally situated accomplishments”, while
Duff’s (2003)hybrid discourse refers to the interweaving of non-academic and academic texts
in a classroom situation.
8
Bayer (1999, 233), in an alternative definition, considers as hybrid those languages that “show a
mix of final and initial heads” with regard to their phrase structure. As this simply implies a choice
between alternatives it would not be considered an instance of hybridity in the model followed
here.
9 Hybridization in Language 145
As we have seen before, the originally English lexicon has undergone a massive
influence from Romance vocabulary. While Old English is still a clearly Germanic
language, Middle English is traditionally delimited by the Norman Conquest as the
starting date for the Romance loans, and Modern English is characterized by its mixed
character (cf. Leisi and Mair 1999, 41–46). This is evidenced particularly nicely in the
following sentence from Decline and Fall of the Roman Empire, in which the few
native elements – all of them grammatical – are italicized (Roberts 1939,37):
The indissoluble connection of civil and ecclesiastical affairs has compelled and
encouraged me to relate the progress, the persecutions,the establishment, the divisions,
the final triumph, and the gradual corruption of Christianity.
According to Scheler (1977, 74), the proportion of loan words in English constitutes
more than 70% of all words. It is only a small step from this extreme intermingling of
two different languages which marked a new period in the history of the English
language to a contact situation that results in a new language spoken by a particular
group, namely to pidgins and creoles, which typically evolve in colonial situations
(cf. Holm 2000, 4–9). A pidgin represents the intermediate step between two distinct
source languages and the creole that develops out of them, which can be regarded as a
language in its own right.
9
This constitutes etymological hybridity par excellence.
When two languages blend to form a new, third one, one of them usually
contributes its grammar (in the case of English, this was the Germanic Anglo-
Saxon), whereas the other contributes its vocabulary (in the case of English, the
Anglo-Norman vocabulary was taken over and many Old English words fell out
of use) (Roberts 1939, 28). Hybrid languages thus do not take over 50% of the
vocabulary and grammar from language A and B each, but they split up the possible
domains of influence (though there may be the occasional A-influence in a B-domain
and vice versa). This raises the question of how important each source’s contribution
to the resulting hybrid language needs to be. Roberts (1939, 28) considers that the
language which contributes its grammar “must be considered the victorious tongue
from the genealogical point of view” and even goes so far as to call it the “surviving
language” (Roberts 1939, 31) at the end of a bilingual situation, which is “colored in
greater or less [sic] degree by the perishing language” (Roberts 1939,31).
10
Even if
one may believe that the contribution of the new language’s lexicon, for example,
should be considered of equal importance to that of its grammar, the question remains
whether the source languages’ contributions necessarily need to have equal weight, as
would be the case in a hybrid organism inheriting 50% of each parent’s genetic
material. The English lexicon, for instance, contains a considerable amount of Ger-
manic words, the Romance influence thus being further diminished in comparison to
the Germanic element that also dominates grammar. Still, English is classified as a
hybrid language. But what if the influence of the second language is even smaller?
While it seems obvious that the existence of a few foreign words in a language are not
9
But cf. G€
orlach 1986, according to whom not every mixed language is a creole.
10
These metaphors can presumably be regarded as closely linked to their time of utterance.
146 C. Sanchez-Stockhammer
enough to consider it hybrid – otherwise all languages would need to be considered
hybrid, so that the concept would no longer be useful – it is still difficult to determine
where to draw the line. For this reason, it makes sense to posit that hybridity ideally
requires an equal contribution of the sources (i.e. 50% for two sources,33.3% for three
sources etc.), while a process with an extreme imbalance between the sources’
contribution should not be classified as hybrid. Where the line should be drawn,
however, is a matter of dispute and will have to be decided individually.
Furthermore, we must ask ourselves whether the existence of a few individual
foreign words in a language already results in something new. Maybe we should
rather consider this an instance of addition. In this respect, Roberts’ (1939)
gradations of mixture are very convenient: when the two languages are equally
strong, and virtually the entire vocabulary of the grammar-providing language
is swallowed up by the other’s, he speaks of intermixture; where one of the
two languages retains both its grammatical and lexical integrity, and where only a
portion of the other is added, he speaks of admixture (Roberts 1939, 31). Note,
though, that even in these cases, Roberts speaks of hybridness.
This leads to the question of whether the opposite exists, whether there is
actually such a thing as a “pure” language. The generally agreed-upon answer
is no (Roberts 1939, 23). Scheler (1977, 85) believes that there has always been
mixing between different languages due to the contact between different people and
peoples. Still, it is not only on the level of individual languages that we are con-
fronted with the question of purity. Most categories are gradient and have fuzzy
boundaries. To give but one very basic example, the speech sounds that are
discussed in linguistic analyses represent an abstraction from a whole “cloud” of
sounds whose articulation is similar enough to qualify as the same. Similarly, there
are better and worse examples of hybrid entities such as individual hybrid genres –
which is sometimes reflected in the existence of a more or less common designation
for the one but not for the other. The idiosyncrasy of the hybrids seems to prevent
strict categorization; they seem to be better suited for a description by prototype
theory. No-one expresses this dilemma better than Nederveen Pieterse (1995, 55):
A theory of hybridity would be attractive. We are so used to theories that are concerned
with establishing boundaries and demarcations among phenomena – units or processes that
are as neatly as possible set apart from other units or processes – that a theory which instead
would focus on fuzziness and me
´lange, cut-and-mix, crisscross and crossover, might well
be a relief in itself. Yet, ironically, of course, it would have to prove itself by giving as neat
as possible a version of messiness, or an unhybrid categorization of hybridities.
The opposite of liminal space is thus situated at the centre of the prototype.
Prototypically, hybridization combines two very different, pure entities – but of
course this does not prevent the (presumably far more frequent) case that several
entities that are hybrid themselves generate a new hybrid. In contrast to cultural
hybridity issues, which often seem to revolve around the problematic idea of purity
(cf. Stockhammer, this volume), linguistic hybridity need not determine whether
the sources are actually pure; one need only determine which features are mixed.
For this reason, from a linguistic point of view nothing prevents hybridization from
hybrids and other entities, e.g. in decreolization processes: when creoles come into
9 Hybridization in Language 147
contact with the source language of the creole’s vocabulary (e.g. in Guyana),
intermediate hybrid varieties may be created from the combination of the hybrid
creole and the creole’s lexical source language (Croft 2000: 213).
Languages also influence each other in the case of code-switching, which society
often seems to regard as a defective and inferior language use (cf. R€
athzel 1999,213
with regard to Turkish teenagers in Germany). While code-switching may be used as a
stylistic means to demonstrate linguistic skills or to play with language, learners of a
foreign language who have not yet achieved full command of it sometimes fill gaps in
the L2 system (vocabulary or grammar) with elements from their L1 (or other
languages). The result of this process, the so-called interlanguage (cf. Selinker
1972), is a hybrid form of language, too. However, the hybrid languages resulting
from such individual shortcomings are to a certain degree idiosyncratic, and the
interlanguage may progress towards a more standard version of the L2 within a fairly
short period of time (or decay, conversely, if there is not enough training). This points
towards another feature of hybridization in language: typically, we would want to
consider not idiosyncratic, ephemeral phenomena but rather those that achieve a more
durable status within the language (and thus the linguistic community).
So far we have considered languages in the sense of (inter)national means for
communication, but not only these may be affected by hybridization processes. The
same phenomena can be observed where dialects of one particular language meet in
a geographical area. This will lead to an overlap of certain features in the local
dialects of the border region
11
– and thus to hybrid varieties as well.
To sum up, we can recognize a gradience in the strength of etymological
hybridity on the level of languages going from idiosyncratic code-switching at
the bottom, which leaves the source languages’ integrity untouched (Weißk€
oppel
2005, 334), to the creation of whole new creole languages at the top.
9.2.7 The Level of Communication
So far, we have basically treated language like an object or an organism whose
hybridity we can investigate. What one must not forget, though, is that language is
no abstract entity but a means of communication, by which human subjects interact
with each other via a channel. Speech events involve the following constituent
factors (based on Jakobson 1985, 150 but with a slightly modified terminology): a
11
Cf. de Saussure (1916/2005, 277–280): where many isoglosses – i.e. geographical borderlines
between dialectal features – coincide, the two areas that are separated constitute different dialects.
However, not all neighbouring languages/dialects are separated by hybrid intermediate varieties.
Thus, there is no intermediate step between the Germanic and the Slavic languages (de Saussure
1916/2005, 280). This is in line with the idea reported above, that languages behave to a certain
extent like biological organisms, so that no hybrid offspring are possible at all between individuals
of very different species (such as dogs and bears).
148 C. Sanchez-Stockhammer
sender sends a message with a particular subject, written in a particular code, to a
receiver via a channel in a particular communicative situation. These various
factors involved in communication are all potentially hybrid. Some of these we
have already considered. Thus, the code refers to language (in terms of sounds,
words, syntax etc.), while the message covers the level of text and text types. As the
subject refers to extra-linguistic events and entities, it is not of interest in the scope
of the present paper – in contrast to sender, receiver and channel.
9.2.7.1 Sender and Receiver
Both sender and receiver of speech are (usually) humans. This means that as soon as
we move away from the level of language as a code, a vast range of options for
hybridity opens up: for instance, we may argue that the linguistic proficiency of each
speaker constitutes a hybrid entity. After all, many micro-systems are mixed in it;
systems that are appropriate for particular communicative situations such as a written
scientific journal article or an intimate spoken conversation among friends. This does
not represent a choice between different alternatives, but rather an intricately interwo-
ven system, in which the parts influence each other. The sender may alsohave a hybrid
identity – either because of a mixed cultural background or simply in the wider sense
because every person takes on different social roles in different situations. This hybrid
identity may have an effect on that person’s particular style of speaking or writing etc.
If a multifaceted linguistic background, e.g. knowledge of several languages or
dialects, is added, this may influence the pronunciation, vocabulary etc., too. Conse-
quently, it is not surprising that Croft (2000: 209) should regard bi- or multilngual
individuals as hybrids. In a very artificial situation, it is even possible to combine the
recorded voices of several speakers and thus to achieve hybrid spoken syllables (cf.
Whiteside and Rixon 2000,935).
Sometimes, the identity of the sender may be unclear, namely when another’s
speech is introduced in the discourse of a novel’s author without any formal markers
(Bakhtin 1981, 303). This double-voiced discourse or heteroglossia,accordingto
Bakhtin (1981, 324), is created by hybrid constructions, which are “an utterance that
belongs, by its grammatical (syntactic) and compositional markers, to a single
speaker, but that actually contains mixed within it two utterances, two speech
manners, two styles, two ‘languages,’ two semantic and axiological belief systems”
(Bakhtin 1981, 304). Even one and the same word may thus simultaneously belong to
two different utterances (Bakhtin 1981, 305). Consider the sentence
It began to be widely understood that one who had done society the admirable service
of making so much money out of it, could not be suffered to remain a commoner
from Charles Dickens’ Little Dorrit: it contains a very typical hybrid construction,
in which the subordinate clause “of making so much money out of it” is in direct
authorial speech, while the main clause reflects a different and more general opinion
about the person referred to (Bakhtin 1981, 306). We thus read a second story, the
author’s story, behind the narrator’s story at each moment (Bakhtin 1981,314).
9 Hybridization in Language 149
The two voices constantly interact with each other as if in a dialogue (Bakhtin 1981,
324). This “mixture of two social languages within the limits of a single utterance,
an encounter, within the arena of an utterance, between two different linguistic
consciousnesses, separated from one another by epoch, by social differentiation or
by some other factor” are called (deliberate, conscious) hybridization by Bakhtin
(1981, 358). He sets off this fairly narrow definition against unintentional, uncon-
scious hybridization, which he believes to be “one of the most important modes in
the historical life and evolution of all languages” (Bakhtin 1981, 358). This latter
concept of hybridization seems closer to the one discussed here – though it is highly
questionable whether the hybridization that results in language changes is always
unconscious –, but it is difficult to compare the two more closely, as Bakhtin does
not define the concept in more detail.
In addition, more than one individual may actually be involved in a text’s
creation: if several people jointly write a text with which they all agree and in
which the individual contributions cannot be distinguished from each other, we can
speak of a hybrid sender, a form of communicative hybridity. When a translator
translates a text, we may even ask ourselves whether we are dealing with two senders
(or with only one – plus the translator as an intermediary black box; cf. Reiß and
Vermeer 1991, 41) – and if so, whether this is an instance of a hybrid sender, too.
Another aspect to consider are quotations or the transmission of others’ ideas in
everyday conversation. Bakhtin (1981, 338–339) believes that the modified repro-
duction of other’s talk constitutes an extremely large proportion of every person’s
utterances and that “re-telling a text in one’s own words is to a certain extent a
double-voiced narration of another’s words” (Bakhtin 1981, 341). Following this
view would mean that half of spoken discourse is hybrid from a communicative
point of view.
As soon as we include the receiver as well, it gets even more complex:
Weißk€
oppel (2005, 325–326) points out that meaning in speech acts always
emerges from the interaction between the participants. To take this one step further,
we may even posit more sources of meaning in communicative situations, among
them the context, the speaker’s intention, the hearer’s interpretation etc. Thus, the
emergence of meaning in communicative situations is extremely hybrid. This is
expressed by Bhabha (1994, 36) in his concept of the so-called “third space”:
The pact of interpretation is never simply an act of communication between the I and the
You designated in the statement. The production of meaning requires that these two places
be mobilized in a passage through a Third Space, which represents both the general
conditions of language and the specific implication of the utterance in a performative and
institutional strategy of which it cannot ‘in itself’ be conscious. What this unconscious
relation introduces is an ambivalence in the act of interpretation.
A similar argument is advanced by Bakhtin (1981, 279):
On all its various routes toward the object, in all its directions, the word encounters an alien
word and cannot help encountering it in a living, tension-filled interaction.
He speaks of the “internal dialogism of the word” (Bakhtin 1981, 279) and
stresses that words should not be looked at in isolation. Bakhtin (1981, 279) also
150 C. Sanchez-Stockhammer
emphasizes the role of the listener (and his answer), at whom (and at whose
conceptual system) the word is directed – and basically the intra- and extralinguistic
context in which words attain their meaning (Bakhtin 1981, 281–282). If we thus
consider that many different sources contribute to a word’s meaning, we may even
argue that words are always hybrid. However, as their meanings are disambiguated
by the context, they are not hybrid from a semantic point of view.
12
Instead, we
have to recognize yet another type of hybridity, namely communicative hybridity.
9.2.7.2 Channel
The communicative channel can be hybrid, too. Thus, spoken and written Japanese
used to be so different from each other that they could be considered separate
systems. Toward the end of the nineteenth century, a new written language, which
adapted spoken elements, emerged from the gembunitchi movement.
13
In recent
years, new electronic means of communication have been blurring the traditional
dichotomy between the spoken and the written modality not only in English but
presumably in all languages making use of them (cf. Brown and Yule 1983, 12–19):
thus Fandrych (2007, 148) advocates regarding electronic forms of communication
(such as text messages, e-mail or chatrooms) as a new medium between the tradi-
tional two media – or, to apply the terminology of the present paper, as a hybrid
medium. Crystal (2001, 48), however, would oppose this view because, among
other things, electronic texts are simultaneously available on an indefinite number
of machines, because they do not degrade when copied, and because they have
permeable boundaries as far as links to others etc. are concerned. In the approach
followed here, though, these other, additional characteristics do not prevent hybrid-
ity. We therefore agree with Fandrych (2007, 152), who suggests that we should
consider media as a scale or cline: spoken – electronic – written – and not as a
separate medium, like Crystal (2001, 238).
There are also certain text types/art forms that transgress media boundaries
(cf. Hansen-L€
ove 1983). A case in point are comics, in which language and pictures
interact in such a way that one is very often needed for the interpretation of the other.
Internet pages may consist of a video and text that refers back to the video; in this
case, the page as a whole is also a hybrid kind of text. Harpold (2005, 110) points out
that some texts are crafted with one medium in mind and consumed in another – e.g.
digital versions of printed texts – whereas others are created with multiple media in
12
According to Bakhtin (1981, 305), it is a frequent phenomenon “that even one and the same
word will belong simultaneously to two languages, two belief systems that intersect in a hybrid
construction – and, consequently, the word has two different meanings, two accents...”. If a word
has different meanings, this does not yet make it hybrid – but if two or more of these meanings are
simultaneously realized in a particular communicative situation, this is hybrid indeed.
13
I would like to thank Noriyo Hoozawa-Arkenau (Heidelberg University) for pointing this out
to me.
9 Hybridization in Language 151
mind, e.g. magazine articles that appear both in print and on the magazine’s website.
Even more extreme in this respect are modern dictionaries – which are now crafted
in a format that can be used in the form of digital applications for computers, mobile
phones etc. Yet I would not consider these examples of hybridity but rather argue
that such texts are underdetermined with respect to their medium.
9.2.8 The Level of Abstraction: Models of Language
We have now covered the various levels of language – but there is still another level
beyond language itself, which is related to it: the abstract level. Linguists attempt to
explain how language works, both in a general sense and with relation to how
humans learn and understand language. The models devised to this end often fail
to take into account certain features, which are then pointed out by subsequent
research. This means that there are usually several competing models which all
have their shortcomings. In order to overcome these drawbacks, there is a tendency
to combine existing models into new models in the next step of the research (cf. e.g.
Stevenson 1994 for syntactic disambiguation, Shen 1999 for metaphor interpreta-
tion and Caillies and Butcher 2007 for the processing of idiomatic expressions).
Since additive models are unlikely to exist, the new models are usually hybrid, so
that there is a lot of language-related hybridization on this level.
14
9.3 Summary and Conclusion
To sum up, language and linguistic elements may be hybrid with regard to the
following variables:
• Formal hybridity (e.g. blends)
• Semantic hybridity (e.g. idioms)
• Functional hybridity (e.g. syntax)
• Etymological hybridity (e.g. languages)
• Communicative hybridity (e.g. double voice in the novel).
This overview of hybridisation phenomena within language is of course not
exhaustive. Even if it had attempted to cover all relevant areas, it is highly probable
that many other language-related issues will be found to be legitimate instances of
hybridity or hybridization, depending on the precise definition that is chosen.
14
Sometimes, though, hybrid is not defined in articles whose title contains this term. In such cases,
it is usually doubtful whether we are actually dealing with hybrid phenomena, and one may argue
that Meskill and Anthony’s (2005) “hybrid” language class is simply instructed with a mixture of
different methods, and that the elements in Watson et al.’s (2003) “hybrid” lists do not interact to
create something other than an additive entity.
152 C. Sanchez-Stockhammer
What we have not considered so far is why hybridization happens in the first
place and who is responsible for it. For even if we have mainly considered language
as a structure, the changes in language are brought about by its speakers, and we can
assume that more often than not, they have a reason why they apply processes of
hybridization. For instance, hybrids may be formed in order to enlarge the range of
options: new words meet the demand to express a newly-emerged concept (and may
be etymologically hybrid if one of the bases has been borrowed from another
language); a new genre meets the demand of a particular author’s wish to express
themselves in a particular way etc. Particularly where the mixed character of the
linguistic entity is very evident, hybridization can also serve as a stylistic means,
e.g. to create a poetic effect. Thus, Dadaist artists such as Hans Arp, Walter Serner
and Tristan Tzara switched between different languages to create a poetic multilin-
gual text (Mersmann 2002). The combination of very disparate constituents may
also result in a comic effect, e.g. in the so-called macaronic texts, which usually
integrate native (e.g. English) words into a Latin system of grammar and inflection,
as in “Boyibus kissibus sweet girliorum, girlibus likibus, askum for morum”
(Hansen 1961, 56). This comic effect can also be achieved with creative word
formations in an otherwise English text, as long as the disparity between the
constituents is felt, e.g. in trade-ocracy or weatherology (cf. Hansen 1961).
While these last examples may evoke the impression that hybridizations are
typically produced by artists, writers, journalists and other creative language users,
this is not entirely true. Even if formal hybrids such as new blends are presumably
formed mainly by these groups, new words in general, particularly compounds, are
often created very spontaneously by all kinds of speaker, so that semantic hybrids
are actually very common among very different acting subjects. Once originally
foreign linguistic elements have become part of the vocabulary of a language, they
may enter these word-formation processes very naturally as well, so that etymolog-
ical hybrids may be formed by the unaware language user.
15
Nonetheless, it seems that hybridization in language very often refers to deliber-
ate new creations rather than to instances of incomplete appropriation, as is
frequently the case in other cultural domains (cf. Werbner 1997, 5, who speaks of
“routine cultural borrowings and appropriations”). Similarly – and in contrast to
cultural hybrids –, the number and amount of influence of the sources of linguistic
hybrids can usually be determined quite clearly, at least on the lower levels of the
pyramid (cf. Fig. 9.1).
16
This may explain a certain tendency for linguistic hybridity
15
However, as long as there is still an exotic element about them, only certain groups of language
users may produce new hybrids. Wehrle (1935, 6) believes that at least the early etymological
hybrids will have been formed in written language (because of their learned character) and entered
the spoken language from there.
16
By contrast, hybrids in language can often not be unique in the sense that they are in other
disciplines because of the distinction between langue and parole (cf. de Saussure 1916/2005,
30–31). Thus the hybrid, e.g. an etymologically mixed word, may be unique on the level of the
system of langue, but not on the level of parole, because it may be used many times by different
speakers in different situations.
9 Hybridization in Language 153
to be likened to biological hybridity.
17
The higher we move up, the more complex it
gets, because language functions as a means of communication within culture – so
that culture may even be added as an additional level at the top of Fig. 9.1.
But let us return to the initial question: is hybridization hybrid? Let us consider
the etymology first (cf. the etymological information from the Shorter Oxford
English Dictionary summarized in Table 9.1):
One constituent comes directly from Latin and the other two have a French
origin. As they all involve Latin at one stage in their history, all three components
could be categorized as having a classical origin. If we decide to look at the parent
generation only, though, we do find a certain degree of etymological hybridity
(Latin and French). From a formal point of view, the final <e>of the second
constituent is deleted – but since this is rule-driven, we do not regard this as an
instance of formal hybridization. Let us therefore consider semantics, which usually
involves some idiomatic element: as long as we determine precisely what is meant
by a hybrid, the derivation itself is fairly regular, with -ize contributing ‘bring or
come into some specified state’ and -ation forming a noun ‘denoting verbal action
or an instance of it, or a resulting state or thing’ (cf. SOED). Hybridization is thus
not a semantically hybrid word either – and therefore only a very weak instance of
hybridization if we consider all levels jointly.
However, as we have seen, hybridization as a phenomenon occurs on all levels
of language. Therefore, if we consider all the instances of hybridization introduced
above, we must necessarily come to the conclusion that language is a hybrid system
par excellence.
Acknowledgements I would like to thank Philipp Stockhammer and Ingrid Fandrych for our
valuable and stimulating discussions about hybridization.
References
Ackermann, Andreas. 2004. “Das Eigene und das Fremde: Hybridit€
at, Vielfalt und Kulturtransfers.”
In Handbuch der Kulturwissenschaften,3rdvol.Themen und Tendenzen, edited by Friedrich
Jaeger and J€
orn R€
usen, 139–154. Stuttgart: Metzler.
Bakhtin, Mikhail. 1981. TheDialogicImagination.FourEssays.Austin: University of Texas Press.
Bartsch, Sabine. 2004. Structural and Functional Properties of Collocations in English.T
€
ubingen:
Narr.
Table 9.1 The etymology of hybridization
Constituent Hybrid -ize -ation
Etymology Latin French <Latin <Greek French <Latin
Meaning ‘X’ ‘turn into’ ‘result of’
17
Note, though, that linguistic hybrids are often fertile, e.g. hybrid words that engender new word
formations.
154 C. Sanchez-Stockhammer
Bauer, Laurie. 1983. English Word-Formation. Cambridge: Cambridge University Press.
Bayer, Josef. 1999. “Final complementizers in hybrid languages.” Journal of Linguistics 35:
233–271.
Bhabha, Homi. 1994. The Location of Culture. London: Routledge.
Blech, J€
org, Rafaela von Bredow, and Johann Grolle. 2005. “Darwins Werk, Gottes Beitrag.” Der
Spiegel 52/2005: 136–147.
Bolinger, Dwight. 1975. Aspects of Language.2
nd
ed. New York: Harcourt Brace Jovanovich.
Brown, Gillian, and George Yule. 1983. Discourse Analysis. Cambridge: Cambridge University
Press.
Bucakli, O
¨zkan, and Julia Reuter. 2004. “Bedingungen und Grenzen der Hybridisierung.” In
PostModerne Dekonstruktionen. Ethik, Politik und Kultur, edited by Susanne Kollmann and
Kathrin Sch€
odel, 171–182. M€
unster: LIT Verlag.
Bußmann, Hadumod. 1990. Lexikon der Sprachwissenschaft. 2nd ed. Stuttgart: Kr€
oner.
Caillies, Ste
´phanie, and Kirsten Butcher. 2007. “Processing of idiomatic expressions: Evidence for
a new hybrid view.” Metaphor and Symbol 22: 79–108.
Croft, William. 2000. Explaining Language Change: An Evolutionary Approach. Harlow: Pearson.
Cruse, David A. 1986. Lexical Semantics. Cambridge: Cambridge University Press.
Crystal, David. 1995. The Cambridge Encyclopedia of the English Language. Cambridge:
Cambridge University Press.
Crystal, David. 2001. Language and the Internet. Cambridge: Cambridge University Press.
Duff, Patricia. 2003. “Intertextuality and hybrid discourses: The infusion of pop culture in
educational discourse.” Linguistics and Education: An International Research Journal 14:
231–276.
Erjavec, Karmen. 2004. “Beyond advertising and journalism: Hybrid promotional news dis-
course.” Discourse & Society: An International Journal for the Study of Discourse and
Communication in Their Social, Political and Cultural Contexts 15: 553–578.
Fandrych, Ingrid. 2007. “Electronic communication and technical terminology: A rapproche-
ment?” NAWA Journal of Language and Communication 1: 147–158.
Fandrych, Ingrid. 2008. “Submorphemic elements in the formation of acronyms, blends and
clippings.” Lexis – E-Journal in English Lexicology 2: 105–123.
Flavell, Linda, and Roger Flavell. 1992. Dictionary of Idioms and Their Origins. London: Kyle
Cathie.
Fludernik, Monika. 2005. “Hybridity.” In Routledge Encyclopedia of Narrative Theory, edited by
David Herman, Manfred Jahn, and Marie-Laure Ryan, 227–228. London: Routledge.
Galster, Christin. 2005. “Hybrid genres.” In Routledge Encyclopedia of Narrative Theory, edited
by David Herman, Manfred Jahn, and Marie-Laure Ryan, 226–227. London: Routledge.
Gauger, Hans-Martin. 1971. Durchsichtige W€orter: Zur Theorie der Wortbildung. Heidelberg:
Winter.
G€
orlach, Manfred. 1986. “Middle English – a creole?” In Linguistics Across Historical and
Geographical Boundaries 1, edited by Dieter Kastovsky and Aleksander Szwedek, 329–344.
Berlin: de Gruyter.
Gut, Ulrike. 2009. Introduction to English Phonetics and Phonology. Frankfurt/Main: Lang.
Hansen, Klaus. 1961. “Makkaronische Sprachformen – hybride Wortbildungen.” Zeitschrift f€
ur
Anglistik und Amerikanistik 9: 49–64.
Hansen-L€
ove, Aage A. 1983. “Intermedialit€
at und Intertextualit€
at: Probleme der Korrelation von
Wort- und Bildkunst: Am Beispiel der russischen Moderne.” In Dialog der Texte: Hamburger
Kolloquium zur Intertextualit€
at, edited by Wolf Schmid and Wolf-Dieter Stempel, 291–360.
Vienna: Gesellschaft zur F€
orderung slawistischer Studien.
Harpold, Terry. 2005. “Digital narrative.” In Routledge Encyclopedia of Narrative Theory, edited
by David Herman, Manfred Jahn, and Marie-Laure Ryan, 108–112. London: Routledge.
Hausmann, Franz Josef. 1985. “Kollokation im deutschen W€
orterbuch. Ein Beitrag zur Theorie
des lexikographischen Beispiels.” In Lexikographie und Grammatik, edited by Henning
Bergenholtz and Joachim Mugdan, 118–129. T€
ubingen: Niemeyer.
9 Hybridization in Language 155
Hausmann, Franz Josef. 2004. Was sind eigentlich Kollokationen? In Wortverbindungen – mehr
oder weniger fest, edited by Kathrin Steyer, 309–334. Berlin: deGruyter.
Herbst, Thomas, Rita Stoll, and Rudolf Westermayr. 1991. Terminologie der Sprachbeschreibung.
Ein Lernw€orterbuch f€
ur das Anglistikstudium. Ismaning: Hueber.
Holm, John. 2000. An Introduction to Pidginsand Creoles. Cambridge: Cambridge University Press.
Irwin, Mark. 2005. “Rendaku-based lexical hierarchies in Japanese: The behaviour of Sino-
Japanese mononoms in hybrid noun compounds.” Journal of East Asian Linguistics
14: 121–153.
Jamieson, Kathleen Hall, and Karlyn Kohrs Campbell. 1982. “Rhetorical hybrids: Fusions of
generic elements.” Quarterly Journal of Speech 68: 146–157.
Jakobson, Roman. 1985. “Closing statement: Linguistics and poetics.” In Semiotics: An Introduc-
tory Anthology, edited by Robert E. Innis, 147–175. Bloomington: Indiana University Press.
Kamberelis, George. 2001. Producing heteroglossic classroom (micro)cultures through hybrid
discourse practice. Linguistics and Education: An International Research Journal 12: 85–125.
Kastovsky, Dieter. 1982. Wortbildung und Semantik.D
€
usseldorf: Schwann-Bagel.
Lauerbach, Gerda. 2004. Political interviews as a hybrid genre. Text: An Interdisciplinary Journal
for the Study of Discourse 24: 353–397.
Leisi, Ernst, and Christian Mair. 1999. Das heutige Englisch: Wesensz€
uge und Probleme.
Heidelberg: Winter.
Levi, Judith N. 1978. The Syntax and Semantics of Complex Nominals. New York: Academic Press.
Lipka, Leonhard. 2002. English Lexicology: Lexical Structure, Word Semantics & Word-Forma-
tion. T€
ubingen: Narr.
McKnight, George Harley. 1923. Words and their Background. New York: Appleton.
Meredith, Mamie J. 1948. “Hybrids.” American Speech: A Quarterly of Linguistic Usage
23: 302–303.
Mersmann, Birgit. 2002. “Weltanschauungen sind Vokabelmischungen. Deterritorialisierung und
Hybridisierung als Entgrenzungsstrategien der Avantgarde.” TRANS. Internet-Zeitschrift f€
ur
Kulturwissenschaften 13. Accessed 30 December, 2010. http://www.inst.at/trans/13Nr/
mersmann13.htm.
Meskill, Carla, and Natasha Anthony. 2005. “Foreign language learning with CMC: Forms of
online instructional discourse in a hybrid Russian class.” System: An International Journal
of Educational Technology and Applied Linguistics 33: 89–105.
Nederveen Pieterse, Jan. 1995. “Globalization as hybridization.” In Global Modernities, edited by
Mike Featherstone, Scott Lash, and Roland Robertson, 45–68. London: Sage.
Nederveen Pieterse, Jan. 1999. “Globale/lokale Melange: Globalisierung und Kultur – Drei
Paradigmen.” In Gegen-Rassismen, edited by Brigitte Kossek, 167–185. Hamburg: Argument.
Nederveen Pieterse, Jan. 2001. “Hybridity, so what? The anti-hybridity backlash and the riddles of
recognition.” Theory, Culture & Society 18: 219–245.
Plag, Ingo. 2003. Word-Formation in English. Cambridge: Cambridge University Press.
Quirk, Randolph, et al. 1985. A Comprehensive Grammar of the English Language. Harlow:
Longman.
R€
athzel, Nora. 1999. “Hybridit€
at ist die Antwort, aber was war noch mal die Frage?” In Gegen-
Rassismen, edited by Brigitte Kossek, 204–219. Hamburg: Argument.
Reiß, Katharina and Hans J. Vermeer. 1991. Grundlegung einer allgemeinen Translationstheorie.
2nd ed. T€
ubingen: Niemeyer.
Roberts, Murat H. 1939. “The problem of the hybrid language.” Journal of English and Germanic
Philology 38: 23–41.
Saussure, Ferdinand de. 1916/2005. Cours de linguistique ge
´ne
´rale. Paris: Payot.
Schach, Paul. 1948. “Hybrid compounds in Pennsylvania German.” American Speech:
A Quarterly of Linguistic Usage 23: 121–134.
Scheler, Manfred. 1977. Der englische Wortschatz. Berlin: Erich Schmidt.
Sch€
onfelder, Karl-Heinz. 1956. Probleme der V€olker- und Sprachmischung. Halle: Niemeyer.
Selinker, Larry. 1972. “Interlanguage.” International Review of Applied Linguistics 10: 209–241.
156 C. Sanchez-Stockhammer
Shen, Yeshayahu. 1999. “Principles of metaphor interpretation and the notion of ‘domain’:
A proposal for a hybrid model.” Journal of Pragmatics: An Interdisciplinary Journal of
Language Studies 31: 1631–1653.
The Shorter Oxford English Dictionary on CD-ROM. 2002. 5th ed., version 2.0. Oxford: Oxford
University Press.
Stevenson, Suzanne. 1994. “Competition and recency in a hybrid network model of syntactic
disambiguation.” Journal of Psycholinguistic Research 23: 295–322.
Watson, Jason M. et al. 2003. “Creating false memories with hybrid lists of semantic and
phonological associates: Over-additive false memories produced by converging associative
networks.” Journal of Memory and Language 49: 95–118.
Wehrle, Otto. 1935. Die hybriden Wortbildungen des Mittelenglischen (1050–1400): Ein Beitrag
zur englischen Wortgeschichte. Freiburg: Weis, M€
uhlhans & R€
apple.
Weißk€
oppel, Cordula. 2005. ““Hybridit€
at” – die ethnografische Ann€
aherung an ein theoretisches
Konzept.” In Globalisierung im lokalen Kontext: Perspektiven und Konzepte von Handeln in
Afrika, edited by Roman Loimeier, Dieter Neubert, and Cordula Weißk€
oppel, 311–347.
Hamburg: LIT.
Werbner, Pnina. 1997. “Introduction: The dialectics of cultural hybridity.” In Debating Cultural
Hybridity: Multi-Cultural Identities and the Politics of Anti-Racism, edited by Pnina Werbner,
and Tariq Modood, 1–26. London: Zed Books.
Whiteside, Sandra P., and E. Rixon. 2000. “Identification of twins from pure (single speaker) and
hybrid (fused) syllables: An acoustic and perceptual case study.” Perceptual and Motor Skills
91: 933–947.
Wurm, S. A. 1995. “The Silk Road and hybridized languages in North-Western China.” Diogenes
43: 53–62.
9 Hybridization in Language 157
