Silly much? Tracing the spread of a new expressive marker in recent corpora

Patricia Ronan
Fakultät Kulturwissenschaften, Technische Universität Dortmund

Abstract

 

In recent times, a new expressive marker has been developing particularly in new media, and the rise of this marker is increasingly manifest in new, large-scale corpora of media and online language. As such, the rise and spread of this marker illustrates the emergence of neologisms and their spread through new media in the 21st century. The structure in question consists of the modifier much, which is postposed to a semantically mostly negatively connoted adjective, noun or infinitive. Most frequently, the syntactic structure is that of a question. Consider the following examples:

Knowledge of, and research on, this structure is scarce so far. It is considered to be an elliptical, frequently ironic, comment or question, with first use observed for 1978 (OED, s.v. much, III.B.1.h.). In an early linguistic note, Liberman (2010) calls the structure X-much. Gutzmann and Henderson (2015, 2019) investigate this structure, here called “expressive much”, with a formal semantics approach. They find that X-much is colloquial, most typically used in question structures, and that it does not act as a quantifier, but instead has the pragmatic force of an expressive. As such, it is a “shunting” expression, it indicates disdain by adding a negative evaluative attitude to an utterance (Gutzmann & Henderson 2015: 267). Hilpert and Bourgeois (forthcoming) argue that this structure has construction status and is currently undergoing expansion from confrontational towards solidarity and alignment with the addressee.

With awareness of this structure just emerging, a number of research desiderates are arising. While the semantics have been described, what has not been determined are the usage patterns of this structure throughout different genres, quantitative elements of distribution and collocation preferences. Thus, the aim of the current paper is to discuss and quantify the usage patterns of X-much data and to determine possible variation in its structure and meaning, as well as its distribution in different varieties of English. This knowledge will allow for the better understanding of the development of the structure, and provide us with interesting insights concerning its international distribution.

So far, only few examples are visible in older corpora. The historical development of the structure is sketched on the basis of the Corpus of Historical American English, COHA (Davies 2010–) and the Corpus of Contemporary American English, COCA (Davies 2008–). Data illustrating the (near-)contemporary usage of the X-much construction has been collected from the 1.9 billion-word GloWbE corpus, consisting of data from 2012 and 2013 (Davies 2013). The collection is semi-automatic: examples of nouns, adjectives and infinitives + much are collected by means of the search interface of the BYU corpus page and false positives are discarded. Data evaluation is partly qualitative, partly quantitative.

After this brief introduction, we will discuss the current state of knowledge of this structure, then introduce our data and methodology and then introduce findings from the current corpus materials, before discussing these and reaching a conclusion.

2. What is expressive much?

So far there has been little research on the expressive much structure, even though interest is rising as shown by the fact that recent research publications are now being produced (Adams 2014; Gutzmann & Henderson 2015; Hilpert & Bourgeois forthcoming). So far, common knowledge about the structure is informed by the OED (s.v. much), which states that expressive much is:

“colloq. (orig. U.S., freq. ironic). With a preceding adjective, infinitive verb, or noun phrase, forming an elliptical comment or question.

The use was popularized by the film Buffy, the Vampire Slayer and the television series derived from it.”

Further, OED gives the following examples of the structure:

In the following we show how the early example in the film script given in example 4, and example 5 from a popular magazine reflect the development of the structure.

Concerning the origin of the structure, Liberman (2010) initially suggested that it is derived from an abandoned polarity item, as in I don't like him much/ Do you like him much? / *I like him much. In further development, there would  have been ellipsis of part of the question, as in “Did you lose much?” to "Lose much?” "Ten thousand.” Liberman argues that concomitantly, generalization has taken place to contexts where X-much was not previously used as intensifying adjunct.

Adams (2014), in a study of the structure in online-media language, points to the importance of the genre of computer-mediated communication in the spread of this structure. He notes the central importance of evaluative use of the structure, and surprisingly he considers examples of X-much to constitute verdictive speech acts. The adjectives that he finds most frequently with the expressive much structure are jealous, pathetic, insane and awkward. Adams (2014: 178) cites ‘Underdeveloped much?’ as the first attested example from the TV show Saturday Night Live, aired on October, 7th, 1978, and notes the later salience of jealous much? in the movie Heathers in 1989, and of morbid much? in Buffy the Vampire Slayer in 1997. X-much structures with nouns are observed in Tuna much? in 1998. Adams (2014: 178–9) considers the uses of the X-much structure as in-group markers for people who recognize the culture of these television programmes, which are to be understood by in-group members as overall critical, accusatory and dismissive. He furthermore observes that the structure started off in question format, but also admitted declaratives from examples from the year 2000 onward (loc. cit.), and can on occasion also express positive rather than critical evaluation (Adams 2014: 183). He sees it as a particular advantage of the structure that it ‘packs a lot of attitude into the fewest possible words’ (Adams 2014: 184).

The semantics of the structure have further been discussed extensively by Gutzmann and Henderson (2015: 269). They argue that the normal modifier much modifies “expressions of all core lexical categories”. Thus, it could be used with adjectives, as in much sweeter, verbal heads or verb phrases, as in Geez! Overreact much? or Wow. Flatter yourself much?, nouns or noun phrases, as in Cliché much? (loc. cit.: 267–268). In terms of its syntactic distribution, they point out that X-much cannot be embedded or modified and therefore has no external syntax (loc. cit.: 269). Contradicting Liberman (2010), they argue that the structure is not in fact derivable from an original verb phrase plus much as in ((are you) angry much?) as it is not answer seeking. Further, they argue that it is not assertive either, as it cannot answer questions (loc. cit.: 269–270). Instead, X-much is argued to be only used expressively and it adds a speaker’s evaluation of a gradable predicate, and thus is a shunting operator. The referent is said to stand out from similar predicates to large degree (Gutzmann & Henderson 2015: 272). When written, Gutzmann and Henderson observe it to be written with a question mark in order to indicate rising intonation. This is interpreted as a sign of interlocutor-directedness and as a check of the hearer’s commitment set (loc. cit.: 279–81).

Hilpert and Bourgeois (forthcoming) work on both meaning and constructional status of expressive much. Hilpert and Bourgeois argue that it is typically used with a critical or sarcastic meaning. As such, the expressive much structure is argued to have typical constructional characteristics, namely idiosyncrasy and non-compositionality of meaning. They further observe, however, that the structure can also be used in a widening set of contexts, particularly in self-deprecating and meta-textual comments and in bragging. Concomitantly, the structure is argued to be undergoing constructional change towards also seeking solidarity and alignment between speaker and addressee.

But not only researchers are taking a professional interest in the expressive much structure and its use, it is also noted by other language users. Thus, metalinguistic comments by language users indicate not unexpected outbreaks of complaint-traditions towards neologisms, as well as equally expected defences of its use. An illustrative example which the current author noted stems from the discussion of this very topic on a major Irish general discussion website, www.boards.ie, which took place in 2010.

I see a lot of people saying things like "angry much?" on facebook and here. There is another example of it on a thread where someone says "stepbrothers much?". It doesn't make sense and it's annoying. Is in [sic] nerdy internet speak or is it an American/British affectation? Does anyone know the origin of it? Thanks. (Rantyface Registered User 16-May-2010 19:09)

Unnecessary linguistic complaint much? (Superbus 16-May-2010 19:11)

This exchange on an online discussion site is very instructive as it illustrates both the manifestation of negative attitudes to neologisms and an ironic meta-discourse. After introducing what is known about the construction so far and presenting some user attitudes, the remainder of the paper will, after discussion the underlying data and method, give an overview of how and where language users do use this structure.

 

As the aims of this study are to give both qualitative and quantitative overviews of the rise of the expressive much structure, as well as its current structure and distribution, both historical data and near-contemporary corpus materials are drawn upon. One source of data is the Corpus of Contemporary American English, COCA (Davies 2008–). The corpus consists of more than 200,000 texts from the years 1997 to 2017, with a total of more than 560 million words from spoken, as well as from formal and informal written genres. A source that has been used to check the time depth of attestations of the X-much structure is the Corpus of Historical American English, COHA (Davies 2010–). This corpus spans the time period from 1810 to 2010 and consists of more than 400 million words.

For near-contemporary international variation, the GloWbE corpus (Davies 2013) is used as a data source. GloWbE contains 1.9 billion words from twenty different varieties of English (United States, Canada, Great Britain, Ireland, Australia, New Zealand, India, Sri Lanka, Pakistan, Bangladesh, Singapore, Malaysia, Philippines, Hong Kong, South Africa, Nigeria, Ghana, Kenia, Tanzania, Jamaica), which were sampled between 2012 and 2013 (Davies 2013). The corpus is representative of online-use of the varieties of English. It may be criticised that in this corpus, the classification of the data as belonging to a certain variety of English is based on the geographic location of the website that the data is taken from. Potentially this could mean that the geographical location of the website is not identical with the variety of English used by its owner. However, given the assumedly restricted percentage of language users to whom this is likely to apply, as well as the overall usefulness of the resource, the use of the corpus definitely seems justified.

For the COCA, COHA and GloWbE corpora, the search strings [j*] much have been used to find the examples of adjectives followed by much. Further, nouns were searched for with the string [n*] much, and here, too, like in the search for adjectives the manual investigation had to be restricted to higher-frequency items of head-word + much with at least 5 attestations. An exception was made for the search of adjectives in COHA, where a relatively small number (404 attestations) of adjective + much structures were found and all of the examples could be investigated. Otherwise, the restriction to a minimum number of attestations proved necessary as the manual extraction of true positives is very work intensive due to the overall very large number of attestations, with many false positives, such as for example Some people described how they got angry much more easily when drunk (GB G youthhealthtalk.org), which then had to be separated manually from true positives such as 1110834 Angry much? See a specialist (PH G gmanetwork.com). However, the restriction to collocations of much with higher frequency nouns and adjectives particularly in the comparatively smaller COCA and COHA corpora does of course mean that particularly emerging and other ad-hoc formations may have been missed in this analysis and only the more established ones were found.

In COCA, of the 365 types of adjectives followed by much, 8 types were true positives of the X-much construction. In GloWbE, we found 2677 examples of adjectives followed by much. These tokens were then investigated manually. During manual investigation, of all those examples of adjective followed by much that were frequent enough to have a currency of at least 5 attestations, 26 types of collocates in the expressive much structure with adjectives were found with a total of 193 true positive tokens. With nouns, expressive much appeared in 6 tokens in 3 different types, and 2 tokens in two different types of verb phrases were also yielded by the [n*] much query in GloWbE. False positives were discarded. These particularly were cases in which an adjective is followed by much, but which are not members of the target structure, such as he is angry much more often. All the remaining examples were investigated manually.

 

The expressive much structure has been observed with the earliest attestation in the Oxford English dictionary stemming from the New York based television programme Saturday Night Life in 1978 (OED, s.v. much). Slightly later examples are attested from a 1988 American TV drama, Heathers (OED, s.v. much). The early history of the construction, however, is not the focus of the current study, this will be treated elsewhere (Ronan 2018).

 

In the corpora under consideration here, COHA, COCA, GloWbE, earliest attestations can be found from the 2000s onwards. In COHA, still very few collocations are found. Even though 404 attestations of an adjective followed by much are found in COHA, only two types in fact contain tokens of the X-much construction under scrutiny here.

The frequency is thus very low in the COHA materials and corresponds to 0.005 examples per 1 million words. More examples can then predictably be found in more recent materials from the Corpus of Contemporary American English, COCA. In the 560 million words of COCA, we find 365 attestations of an adjective followed by much, and of these 8 types contain 11 examples of the X-much construction (0.0196 examples per 1 million words). These are the adjectives jealous (4 examples), and ambitious, bitter, desperate, dramatic, obsessive, pretentious and uptight. Amongst these examples we also find the two examples that are also attested in COHA, examples number 5. and 6. above. With one exception the attestations in COCA stem from the category of fiction, and are roughly equally distributed across the time-span represented in COCA. The only example that does not yield from fiction stems from a magazine (example 8):

Example 5 represents the earliest attestation of the X-much construction in COCA. No particular rise in numbers is visible during the following decade, with one example each being found in 2004, 2006, 2009, 2013 and 2014, and two examples each in 2007, 2012, 2015.

Given the low numbers of attestations of the structure particularly in COHA, which is based on traditional genres of writing, and the only slightly better attestation in COCA with 0.0196 examples per 1 million words, it seems likely that the development of the structure has not taken place in these traditional genres and currency remained low even as late as in the early 21st century. As early attestations stem from the depiction of spoken language in the media, new media are likely to provide better sources of the structure. Thus, the GloWbE corpus of internet based language is likely to offer a better source for data collection.

 

The GloWbE consists of 1.9 billion words of various genres of online language stemming from the years 2012 and 2013 (Davies 2013). Searching for the string [*j] much, 2677 examples are returned. During manual investigation, of all those examples of adjective followed by much that were frequent enough to have a currency of at least 5 attestations per type were investigated. 26 types of collocates in the expressive much structure were found with a total of 193 true positive tokens. Of those 26 types, 24 contained negatively connoted adjectives, and only two (happy, amazing) had positively connoted adjectives. It is noteworthy that a large part of these examples can be found in users’ comments on the websites in question.

The most frequent expressive much types were the following:

Jealous > bitter > paranoid > angry > excited > desperate > dramatic > crazy

Illustrative examples are given in examples (9) and (10).

It is noteworthy that for the collocations desperate much, jealous much, dramatic much and paranoid much, all the examples were true positive tokens, no false positives were found. Collocation strength therefore will be comparatively high here.

 

According to previous descriptions of the expressive much structure in OED (s.v. much) and in Gutzmann and Henderson (2015, 2019), the structure has been described as having the following properties: it mainly consists of adjectives and nouns which are negatively connoted. It is used in question contexts as indicated in writing by the use of a question mark. Further, it is extra-syntactic and thus shows no syntactic embedding. In the following, these claims are investigated on the basis of corpus materials derived from the GloWbE data.

4.2.1.1 Typical structure of the X-much construction

In the corpus data from GloWbE, the typical structure of the X-much construction is mainly that of short, extra-syntactic question structures as noted by other authors (e.g. Gutzmann & Henderson 2015: 269) and illustrated in examples (11)–(14) below.

The data illustrate that the phrases that are used indeed are mostly short question structures. The modified adjective or noun characteristically consists of a short, typically one-word head plus much. Typically, the head is also semantically negatively connoted. For some adjectives it could be observed that all the collocations of this adjective with much were in fact true positives of the X-much construction, while for other adjectives, only some of the co-occurrences of the adjective with much were in fact examples of the X-much construction. Which types have the highest incidence of the X-much structure compared to non-X-much can be seen in Table 1 and Figure 1 below.

Table 1 and Figure 1 illustrate that for some adjectives it is considerably more likely that they are true positives of the X-much construction than other adjectives. For the adjectives jealous, paranoid, desperate, dramatic, hypocritical, rude, creepy, obsessed, ignorant and ironic, all the co-occurrences with postposed much were in fact examples of the X-much construction, while in hungry, obvious, sick, tired or empty, only a small percentage of the co-occurrence of these types were tokens of the X-much construction. It is noteworthy that these adjectives which collocate strongly with the X-much construction can all be described as adjectives expressing mental state. It seems possible that this semantic type of adjective, expressing mental state, formed a linguistic prototype from which the construction then extended towards other collocations.

 

As noted by other authors (Adams 2014; Gutzmann & Henderson 2015), and as also argued in 4.2.1.1 above, the X-much construction repeatedly expresses negative value judgements. However, not all the uses of the X-much construction do so. Hilpert and Bourgeois (forthcoming) argue that the X-much construction is extending towards meanings of solidarity and alignment with the addressee. Similarly, data in the GloWbE corpus offers evidence for the extension of the structure to positively connoted adjectives in addition to the prevalent semantically negatively connoted adjectives and nouns, namely the collocations amazing much and happy much, as illustrated in examples 15 from Jamaica and 16 from Singapore below.

Of these positively connoted heads, only the two above types with two tokens each can be found (2 percent of the X-much tokens in GloWbE). These tokens stem from New Zealand, Singapore, Jamaica, and Malaysia. Here we might argue that they are instances of the linguistic creativity of the language users in these areas. While in the case of the example from Jamaica one might still argue that the extension to a positively connoted adjective may be a sign of incomplete control of the construction by the speaker because the structure is generally rare in Jamaica, the example particularly from Singapore comes from a high-usage country. The structure is also not infrequent in New Zealand and Malaysia. This suggests that these expansions to new contexts may be due to the linguistic creativity of the language users. Or indeed, we might be observing the expansion towards alignment functions as argued by Hilpert & Bourgeois (forthcoming). The issue of inter-varietal differences in the use of the construction will be returned to in Section 4.4.

 

Typically, the X-much construction is spoken with a rising intonation in spoken language, and in written language the interactional nature is represented by the use of a question mark (Gutzmann & Henderson 2015: 269, 279). However, as also observed in Adams (2014), examples can be found which are not, or in the case of 17 below not only, marked as overt question structures.  Even though the syntactic form is largely that of a question, examples which are not, or not only, marked with overt question structures can also be found. These structures are not (or in the case of 17 below not only) questions. Example 17 illustrates nicely the perceived status between (rhetorical) question and emphatic statement. In total, there are 27 examples of the X-much construction (14%) which are not, or not only, used in question structures. 24 examples do not show any question marking at all; 3 examples, comprising two examples of the adjectives excited and one ignorant, are marked with both question and exclamation mark.

Such an emphatic statement confirms Gutzmann and Henderson’s (2015) assessment of the structure as conveying the irritation of a shunting operator. Further, the semantic alignment with the interlocutor which Hilpert and Bourgeois (forthcoming) are arguing for is also likely to be expressed in a non-question structure, and this argument is supported by examples (15) and (16) above. An overview of the adjectives used in non-question structures in the X-much construction is presented in Figure 2 below.

Of the total of 27 non-question examples, jealous much is the most frequently used adjective type in non-question format with 8 examples (29.6% of the examples). As jealous much is the most frequent type of adjectival construction (23.3% of all the types) this is not too surprising. Jealous has been identified as the only actively used type from 1989–1997 by Adams (2014: 179). Further, looking at the examples of jealous much, as illustrated by example 19 above, it becomes obvious that these non-question examples are often applied to the speaker themselves. They represent a statement of the speakers about themselves rather than a (rhetorical) question to others. The fact that the speakers are self-assessing explains why the examples are presented as statements rather than in question format. This also holds for example 20, where the speaker self-assesses as being crazy. In total, of the 27 X-much construction in non-question structures, 10 examples (37%) have first person reference, 5 examples (18.5%) have second person reference and 6 examples (22%) each have either third person reference or the reference cannot be determined due to lack of sufficient context in the retrievable source text.

 

Gutzmann and Henderson (2015: 269) argue that the X-much structure is extra-syntactic, that it cannot be modified or syntactically embedded. The study of the material in the GloWbE corpus materials, dating from 2012 and 2013, largely confirms this observation. Typically the language users either use the structure on its own, as in # Maddave: Jealous much# (AU G cyclingtips.com.au), between brackets or hyphens as in Charles of Brisbane - bitter much - wow what a sore loser! (AU G foxsports.com.au), or divided from the main sentence by other means of punctuation, as in gossip and an idea of what goes on during a Fashion Week Event! EXCITED MUCH?!: o) # (IE G sweetiesal.com). However, exceptions are found.

On the one hand, we find X-much being applied to phrases, as in examples 20 and 21 below. Further, the X-much structure modifies phrases in all those 8 examples in the GloWbE corpus where a noun phrase is modified.

In addition to X-much modifying adjectival phrases, we do find examples of the structure in fact being syntactically integrated, as in examples (23) to (27) below.

There are 8 syntactically integrated examples in total amongst the 193 tokens of the adjective-much construction (4.2%). Of these, jealous much is again the best represented with 3 tokens (37.5%), other types appear as single tokens only. As jealous much is the most frequent type, both in the data investigated in Adams (2014) and in the current corpus material, it is possible that the use of jealous much is most strongly entrenched as an X-much structure and that this type is therefore most likely to offer a starting point for innovation. Support for this suggestion is offered by the syntactic flexibility of the structure both in terms of syntactic integration and in terms of semantic extension to referencing first person alignment.

Potentially, example 27 could also be an instance of syntactic integration of an X-much structure, as we could interpret this as an extrasyntactic question structure with a post-modified phrase.

However, example 27 may also just present an elliptical structure derived from Do you dream big much, Dr. Moses?, which would rule this out as an example of the X-much construction.

These examples, particularly the clearly syntactically integrated examples in 20 to 26, provide counterevidence to the claim that X-much cannot be modified or embedded. These examples may therefore be indicative of increasing complexity of the structure and thus increasing grammaticalisation.

 

Generally this survey of data taken from recent international, internet-based usage of expressive much indicates that the descriptions of the X-much structure provided in OED (s.v. much) and by Gutzmann and Henderson (2015) can be confirmed. The structure is used with adjectives and nouns. Use in verb phrases has not been investigated here, but is attested in GloWbE, particularly with non-finite verbs as in Women have huge intellectually challenged egos. " # Have a problem with women much? (NZ G guymcpherson.com). It can also be shown that semantically the heads of the X-much construction are for the most part negatively connoted. At 2% of the overall examples in the corpus, positively connoted adjectives are rare.

Structurally speaking, typically the written structure is followed by a question mark, to indicate question contexts. The X-much structure appears in declarative contexts in 14% of the corpus examples, and it has been shown that this frequently is the case where the X-much structure refers to the speaker themselves. Concerning the previously described lack of syntactic embedding, it can be seen that extensions of the structure to syntactic embedding are visible in 4.2% of the cases.

However, as the data from the GloWbE corpus stems from number of different varieties of English, the possibility should be considered that different varieties may show differing use of the structure. In the following, the corpus data is investigated for intervarietal differences in frequency of this structure’s usage. 

 

Comparing usage in different varieties of English in the GloWbE corpus, it can be found that the distribution of the expressive much structure varies highly significantly in different varieties of English. [2]

Table 2 illustrates that highest attestations per word-count can be found in some, particularly L1, Asian-Pacific varieties of English, Australia, followed by Singapore, the US and the Philippines. Middle usage ranges are found in predominantly British-based, L1 and L2 varieties of English in South Africa, New Zealand, Malaysia, Britain, India, Pakistan, Canada and Ireland. Low usage is observed in contributions from the African based varieties in Jamaica, Ghana, Nigeria and Kenia. Finally, corpus material from some Southeast Asian varieties shows no examples of the expressive much structure at all. These are Sri Lanka, Bangladesh and Hong Kong, as well as the African country of Tanzania. These findings are presented in condensed form below:

Highest usage (> 0.19/mil.): Australia > Singapore > US > Philippines
Mid-usage (> 0.7/mil.): South Africa > NZ > Malaysia > GB > India > Pakistan > Canada > Ireland
Low-usage (> 0/mil.): Ghana > Jamaica > Nigeria > Kenia
Zero-usage: Sri Lanka, Bangladesh, Hong Kong, Tanzania

Thus we can see that the geographic hotspots for the use of expressive much as indicated in the GloWbE data are (parts of) South-East Asia, South-Africa and the US. Low spots in the GloWbE data, by contrast, are central Africa and Hong-Kong. Established, British-oriented L1 and also L2 varieties have middle positions over all.

These quantitative findings on the use of the X-much structure in different varieties of English correlate well with the observed origin of the structure in American English. Having been popularized by an internationally successful TV movie and series, Buffy the Vampire Slayer, it would have been able to spread internationally in English-speaking countries which broadcast the programme, and through interaction on the international posting boards related to the programme. In addition to having been taken up in the United States themselves, the structure can be shown to be particularly well-represented in Pacific regions, Australia, Singapore and the Philippines, and, to a lesser extent, New Zealand. Overall, non-Pacific, Inner Circle, British based varieties of English, Great Britain, Canada and Ireland, but also British English-based Outer Circle varieties such as India, Pakistan and South Africa, seem to have taken to the new expression somewhat less enthusiastically, and the neologism has been criticised as being foreign by some language users as shown in Section 2. Varieties of English based in Africa, other than South African, only show low counts. This could conceivably be due to less extensive Americanization of these varieties, and perhaps cultures, than those of South Africa, with less wide-spread access to the programme from which the X-much structure spread. However, the Jamaican English corpus data also only shows comparatively low counts. That Jamaican English originates from heavily African-English influenced varieties is interesting in this respect. The question may arise whether the presence of pidginized registers in all these varieties plays a role. Concerning the lack of examples from Sri Lanka, Bangladesh, Hong Kong and Tanzania, no definite suggestion can be made as the X-much structure exists in varieties of English in the broad geographical regions. For Sri Lankan and Bangladesh English, it could be speculated that a certain conservativeness of society might play a role, but both synchronic and evidence for the development of the structure would be needed to obtain further insights. However, the argument of conservativeness does not seem likely for Hong Kong English. Thus, further research on why, and whether, these varieties appear to shun the X-much structure would be desirable.

 

The investigation of the data has shown that the use of the X-much structure in the GloWbE corpus materials, dating from 2012 and 2013, largely follows the observations that have been made in a previous non-corpus-based study (Gutzmann & Henderson 2015). The X-much structure is syntactically independent (95% of the attestations), has a question structure in 86% of the corpus examples, and is found in negatively connoted contexts in 98% of the attestations. However, further features can be observed in this corpus-based study.

It can be seen that there is the possibility that the X-much structure can be structurally complex, that not only single words or phrases with one modifier are used as heads of the structure. In addition, some few cases of syntactic integration can be found. The structure may thus be moving away from an extra-syntactic on to an embeddable phrase.

It can further be observed that varieties in which the structure is used with high to mid-frequency also show examples of semantic extension to not only negatively connoted adjectives and nouns as argued by Gutzmann and Henderson (2015), but are also appearing with positively connoted adjectives. As Hilpert and Bourgeois (forthcoming) are arguing for X-much to be a construction that develops towards speaker alignment, the extension to positive contexts would also fit that mould, as would the fact that the X-much construction to a large percentage, about one third of the examples in the corpus, refer to the speakers themselves.

The variationist results, which compare the extent of the usage of the X-much construction in the different varieties of English that contribute to the GloWbE corpus, likewise offer interesting insights. In relation to the corpus sizes of the different varieties represented in GloWbE, which vary greatly, the attestation of the X-much constructions are highest in the Australian component, followed by Singapore and American English. Considering that the structure is thought to have been first attested in American English, it is notable that other varieties have overtaken the frequency of attestation in the American English component. What is particularly noteworthy is that one of the varieties showing highest attestation numbers, Singapore English, is commonly referred to as a second language variety of English (Kachru 1985; Schneider 2007). However, specific research into the multilingual situation in Singapore also shows that large numbers of young speakers acquire English as a first language variety, with other languages being acquired as second and further languages (Buschfeld 2019). Given this highly multilingual setting in Singapore, this may of course increase linguistic creativity in the use of English language structures, which might accelerate the creative use of linguistic structure in the languages in contact.

 

This paper has set out to discuss and quantify the usage patterns of X-much in recent corpus data and to determine possible variation in its use in different genres and varieties of English.

The study, mainly basing itself on data from the GloWbE corpus, has confirmed that the structure is predominantly used with negatively connoted adjectives in syntactically independent question structures. However, in corpora based on the language of computer-mediated communication, the structure also stabilizes in affirmative, non-question uses, apparently partly driven by use in first-person contexts, towards higher syntactic integration and towards positively connoted adjectives, which are likewise found with reference to first person speaker contexts.

Concerning the spread of the structure, a generally strong representation of the X-much construction in Asian Pacific varieties can be shown, along with the presumed source variety of American English, and generally little uptake, so far, in a number of African English varieties represented in the GloWbE corpus. These observations obviously beg the question of why this should be the case. It is suggested tentatively in this paper that this could be due to a lower impact of American features in these varieties. Further research on what features determine the spread of the structure is a definite desiderate. Possibly language attitudes and cultural attitudes might play a role here, and attitude studies might shed further light on the spread of the structure.

Furthermore, it would of course be instructive to trace the further developments of the structure in even more recent and even larger, or more specialized, corpora to see in how the use of the X-much structure develops in the varieties under scrutiny here, and in how far the frequency of its uses might be changing in these varieties, and how it will continue to change over time.

 

[1] Note that # followed by a space in the GloWbE corpus denotes the beginning of a post, not a hashtag in the conventional social-media sense. [Go back up]

[2] Chi-squared: corpus size: positive tokens = 72.845, df = 19, p-value = 3.076e-08. [Go back up]

 

Discussion of X-much at the boards.ie site: https://www.boards.ie/vbulletin/showthread.php?p=65918604

COCA = The Corpus of Contemporary American English. 2008–. Compiled by Mark Davies. https://www.english-corpora.org/coha/

COHA = The Corpus of Historical American English. 2010–. Compiled by Mark Davies. https://www.english-corpora.org/coha/

GloWbE = Corpus of Global Web-Based English. 2013. Compiled by Mark Davies. https://www.english-corpora.org/glowbe/

 

Adams, Michael. 2014. “Slang in new media. A case study”. Global English Slang. Methodologies and Perspectives, ed. by Julie Coleman, 175–186. London: Routledge.

Buschfeld, Sarah. 2019. Investigating Children’s Acquisition of English as a First Language in Singapore: With a Comparison of Data from the UK. London: Routledge.

Gutzmann, Daniel & Robert Henderson. 2015. “Expressive, much?”. Proceedings of Sinn und Bedeutung 19, ed. by Eva Csipak & Hedde Zeijlstra, 266–283. https://www.danielgutzmann.com/work/expressive-much

Gutzmann, Daniel & Robert Henderson. 2019. “Expressive updates, much?” Language 95(1): 107–135. https://www.danielgutzmann.com/work/expressive-updates-much

Hilpert, Martin & Samuel Bourgeois. Forthcoming. “Intersubjectification in constructional change: From confrontation to solidarity in the ‘sarcastic much?’ construction”. Constructions and Frames.

Kachru, Braj B. 1985. “Institutionalized second language varieties”. The English Language Today, ed. by Sidney Greenbaum, 211–226. Oxford: Pergamon Press.

Liberman, Mark. 2010. “X much”. Language Log. https://languagelog.ldc.upenn.edu/nll/?p=2836

OED = Oxford English Dictionary. “much, adj., adv., pron., and n.”. OED Online. December 2018. Oxford University Press. http://www.oed.com/view/Entry/123133

Ronan, Patricia. 2018. “The rise of expressive much structures”. Paper presented at the 20th International Conference on English Historical Linguistics, University of Edinburgh, 27–30 August 2018.

Schneider, Edgar W. 2007. Postcolonial English: Varieties around the World. Cambridge: Cambridge University Press.