On Euphemisms, Linguistic Creativity, and Humor

The authors would like to thank PALA conference goers for their participation in this project at the University of Kent, as well as two anonymous peer reviewers for their feedback and suggestions. We thank one reviewer in particular for the ‘lipstick lesbian’ example. “The art of euphemism — refusing to use painful words like ‘dying’ — has not passed away.”


Introduction 1
Euphemism may seem over the hill because it is a rhetorical figure with a long history (Horak [2009: 85]).In Ancient Greece, Aristotle [1457b] was one of the first to discuss such figures of substitution in Poetics (Kennedy [1991: 295]); and euphemisms have been used ever since.In recent years, researchers' interest in euphemism has been gradually increasing.For instance, a search for publications about "euphemism" in the Science Direct database from Elsevier, and in the JSTOR database, reveals there has been a steady stream of publications on the topic for several decades, with the pace apparently picking up in the past few decades.So interest in euphemism seems to be increasing (Allan [2019]).And yet, there sometimes seems to be a lack of consensus when linguists define euphemism (Gomez [2009: 725]).This is common in social science, where "the warring triangles" in academic discourse manufacture disagreements that can, in turn, fuel research projects (Martin [2014: 3]).

2
So what is a euphemism?According to the Concise Oxford Dictionary of English Etymology, the word "euphemism" entered English in the 17 th century.Its roots are in the Greek words "euphēmismós […] [to] speak fair and eúphēmos [or] fair of speech," with the adjective "euphemistic" entering English usage in the 19 th century (Hoad [2003]).The Greek prefix eu-means good, which may be why Allan & Burridge [2006: 29] define euphemism as "sweet talking," in contrast to dysphemism ("unfavorable speech") and orthophemism ("straight, neutral speech").While Allan [2012: 1] used the superordinate term "X-phemism" to encompass these figures, Allan & Burridge [2006: 33] note that "a euphemism is typically more colloquial and figurative (or indirect) than the corresponding orthophemism."In other words, forms of X-phemism may occupy different locations on clines for register (from informal to formal) or directness (from literal to figurative).
For centuries, scholars have made taxonomies of rhetorical figures, and euphemism is no exception to this rule.For example, Burridge [2012] recently proposed a useful taxonomy of six types of euphemism.First, a "protective euphemism" is used "to shield and to avoid offense," especially when talking about taboo topics (Burridge [2012: 67]).According to Keyes [2010: 2], "Euphemisms represent a flight to comfort, a way to reduce tension when conversing.They are comfort words."Using euphemisms to discuss taboo topics or unpleasant subjects is very common.Following Frazer [1911], Allan & Burridge [1991: 11] noted that a euphemism can be "used as an alternative to a dispreferred expression, in order to avoid possible loss of face: either one's own face or, through giving offence, that of the audience, or some third party", (quoted in Jamet [2018: 1]).As Allan [2012: 1] states elsewhere: Where the taboo is very strong, and/or one or more of the interlocutors has a subjective emotional engagement with the topic, euphemism is preferred because it focuses away from the (potentially) offensive.
This highlights the conscious and intentional use of euphemisms based on social awareness.For instance, euphemisms for death, such as pass away, may be more apt to use in some settings, while speakers in other settings might use rude idioms, such as kick the bucket or bite the dust instead.As Keyes [2010: 1] puts it, "we all rely on euphemisms to tiptoe around what makes us uneasy and have done so for most of recorded history."This is true of protective euphemisms.
The second type for Burridge, the "underhand euphemism," is used "to mystify and to misrepresent" something dishonestly, as in Orwellian Newspeak, and politicians and criminals are just as likely to use them (Burridge [2012: 68]).To sugarcoat bad news, for example, managing directors may use underhanded euphemisms when making public announcements about upcoming redundancies because firing staff is unpopular and unpleasant.Similar euphemisms may also be used by human resource managers in charge of actually carrying out orders to give staff the pink slip.In such contexts, underhand euphemisms such as payroll adjustment (Holder [2007: 293]) might be more acceptable than fire or dismiss or sack, even if critics chastise officials for their doublespeak and underhand euphemisms.
Third, an "uplifting euphemism" can be used "to talk up and to inflate" something, as occurs in technical, legal, or bureaucratic jargon sometimes (Burridge [2012: 69]).For example, it is common for car dealerships in America now to market used cars as preowned cars, as "ownership" has a better connotation than "use".One could also imagine a modern-day Robin Hood calling his job revenue redistribution.When Crystal [1994: 172] refers to euphemism as the "use of a vague or indirect expression in place of one which is thought to be unpleasant, embarrassing, or offensive," he suggests that an uplifting euphemism which "is thought to be […] offensive" now could have been widely accepted before.For instance, as Halmari [2010: 829] shows, euphemisms in the USA for people who were feebleminded or mentally retarded were widely used by officials several decades ago.But they eventually became offensive, and were replaced by other euphemisms, such as intellectually disabled.Thus, uplifting euphemisms may be replaced by others when future generations see them as offensive.
Fourth, a "provocative euphemism" can "reveal and [...] inspire", as some politically correct euphemisms may demonstrate (Burridge [2012: 70]).In Switzerland or Germany, for instance, it is very common to officially refer to certain people as having migration background, much as the term visible minorities is officially used in Canada.Burridge [2012: 70] compares terms like Italian American and Japanese American to African American.One might add Native American (Holder [2007: 270]) or first people (Holder [2007: 180]) or First Nations here, too.These may be examples of people selecting the terms they want others to use to describe them.But as their usage grows, such terms may seem less provocative over time.In their study, for instance, Pinker et al. [2008] offered at least two reasons why we sometimes prefer to use indirect language over direct language.The first is "plausible deniability" (e.g.you can claim your offer of a bribe was misunderstood by the cop); and the second is "negotiating social relations" (e.g. relations of dominance or sharing or reciprocity), which politeness can help us achieve.Thus, a euphemism may be provocative yet polite, and help us "negotiate social relations."Fifth, a "cohesive euphemism" helps "show solidarity and [..] define the gang," as a form of insider language seen, for example, among hospital staff faced with death and disease every day (Burridge [2012: 70-71]).Informal anecdotes abound of hospital doctors who use euphemisms to speak politely to patients and their family members, only to use crude dysphemisms about their patients when they are amongst colleagues in the staff room.In fact, as Casas Gómez noted [2009: 738], euphemism can distance or "attenuate […] a certain forbidden concept or reality," while dysphemism can "reinforce" it.When used in these intimate workplace settings, away from public ears, crude dysphemisms may function as cohesive euphemisms for groups of co-workers.
Finally, a "ludic euphemism" can be used "to have fun and to entertain" people in playful and creative contexts (Burridge [2012: 71]).For instance, as Crespo-Fernández ([2015: 47)] claims, ludic euphemisms are good for "diffusing the seriousness of taboo subjects."Indeed, tension reduction is an important pragmatic effect (Colston [2015: 81]), and ludic euphemisms may be used for humor to achieve that effect.Euphemisms for bodily functions or sex offer many instances of this type.For example, Holder [2007: 105] lists funny instances such as break wind for "fart", night games for "copulation" [2007: 275], and plumbing for "the parts of the body concerned with defecation and urination" [2007: 302].Indeed, these topics are rich sources of examples, with euphemisms for sex comprising the largest category in Holder's dictionary.We will discuss this sixth category in more detail later.Indeed, a euphemism can even turn into a dysphemism, which may be why Keyes [2010: 1] states: "Euphemisms are a function of their times."As our times change, so do our euphemisms, as Halmari's study demonstrates [2010].The most vivid image of this process comes from Pinker, who memorably called it "the euphemism treadmill" [1994: A21]: one euphemism replaces another as our sensitivity changes.
But what about the very process of creating euphemisms?This is where our study comes in.Here, we present the results of a study in which two dozen subjects produced over 60 euphemisms with novel definitions in response to a euphemism creation task.Although examples like intellectually challenged are compounds composed of an adverb and an adjective (based on a participle), our study focuses on euphemisms composed of nominal compounds, where the first noun (N 1 ) often functions as an adjective for the second (head) noun (N 2 ).The two questions this project aimed to answer were as follows: 1. How creative can people be when asked to produce a few novel euphemisms in a very short time with the same starting conditions (a list of 36 nouns)? 2. Is it true that the more neutral the meaning of both N 1 and N 2 , the more likely it is for the Noun-Noun Compound (NNC) to be used in doublespeak?
In the sections that follow, we first discuss some of the relevant research on nominal compounds, before describing our study and the results in more detail.We then discuss the results in relation to humor, and conclude with a summary and suggestions for future research.

Nominal compounds and euphemisms
There is a wealth of research on nominal compounds that a brief survey like this cannot discuss in full due to limitations of space and time.Yet several studies merit some discussion.This is because the euphemisms we study are nominal compounds.To begin, Bauer [1998] observed two trends amongst linguists doing research on nominal constructions, which involved either splitting apart the forms into two groups, or lumping them together into one category.According to Bauer [1998: 65]: The splitters see two classes of noun + noun sequence in English: syntactic constructions consisting of nouns with nominal modifiers, and compounds.The lumpers see a single class […] of compounds.
While words like girl friend have alternative spellings, such as girl-friend or girlfriend, it does not mean that two words have been definitively fused into one (Bauer [1998: 69]).This poses problems for those who hold that compounds are spelled as one word.Stress is another concern since the rule that compounds have initial stress does not always hold.Those who say that ˈapple cake is a compound, while apple ˈpie is a syntactic construction" overlook the fact that the same words can be stressed differently at different times (Bauer [1998: 70]).Other criteria for making distinctions between compounds and syntactic constructions pose similar problems in Bauer's opinion.That is why he concludes by siding with the lumpers, and basically defends the idea of there being a single category, until proven otherwise (Bauer [1998: 70]).
In Geeraerts' recent discussion of Flemish noun + noun compounds, such as schapenkop (lit.sheep head / fig.dumb person) and droogkloot (lit.dry testicle / fig.boring person), he proposed a model for their interpretation.His "Prismatic Model" (Geeraerts [2002: 466]) proposes that listeners can quickly move through six steps of On Euphemisms, Linguistic Creativity, and Humor Lexis, 17 | 2021 interpretation to reach the figurative meaning of the compound, when that meaning is the intended meaning, either for the purposes of insult and injury, or for the purposes of humor.As Geeraerts [2002] sees it, after a literal reading of the whole, followed by a literal reading of the parts, there is a figurative reading of the whole, reinforced by a figurative reading of the parts.While context of course plays a role in this interpretative procedure, moving from the whole to the parts in successive stages seems a plausible process of interpretation.
Frequency of usage may also play a role in interpretive processes.Although Maguire & Cater [2005] found no experimental evidence to support Gagné & Shoben's [1997] competition among relations in nominals theory (CARIN), they did agree that "combinations that are encountered very frequently can be stored as single entries in the lexicon, thereby obviating the combination process" (Maguire & Cater [2005: 111-112]).This finding, which may support Bauer's view of compounds working as coherent units, adds to our understanding of how nominal compounds are interpreted.More recent research by Smith et al [2014] brings the lumpers and the splitters even closer together.In a nominal compound (NNC) such as mouse mat or snow smoothie, Smith et al [2014: 100] write that [the] only invariant information deducible from the structure of the NNC itself is that it denotes something (conveyed by the head [noun]) that is somehow related to something else (conveyed by the modifier).
Because Smith et al [2014: 135] found NNCs to be "products of the interaction of semantics and pragmatics" -i.e. of "conventional meaning ('code') […] [and] "reasoning ('inference')" (italics in the original) -their finding adds support to the conclusion reached earlier by Bauer about treating combinations as whole units.
Other recent research on euphemisms relates to frequency of usage as well.Interestingly, McGlone et al [2006] did not find evidence to support the "associative contamination hypothesis" -another term for Pinker's "euphemism treadmill."The hypothesis of associative contamination implies that as terms become contaminated or take on negative connotations, euphemisms with new terms replace them.Instead, what McGlone et al found was experimental evidence to support their "camouflage hypothesis," suggesting that the more conventional a euphemism is, the more effective it is.This is because it is "camouflaged;" in other words, it attracts little attention, is easy to process, and is thus effective.The amount of "camouflage" a euphemism has increases with its age, such that heed nature's call (urinate) has more "camouflage" than make a pit stop because the first one is several centuries old in English, while the second is just a few decades old (McGlone et al [2006]).The humoristic value of a highly camouflaged euphemism, though, may be limited.If it is too subtle, and not recognized consciously, then nobody may find it funny.
Turning specifically to euphemisms in nominal compounds, in her study, Portero Muňoz [2011] analyzed dozens of euphemisms from the 2008 financial crisis.She discussed examples such as employment gap -a "gap caused by a lack of employment" (Portero Muňoz [2011: 145]) in someone's CV.While there is never a gap in someone's existence, there may be gaps in someone's career.Because such a period might look bad on a job seeker's CV, the word unemployment is the taboo topic to be avoided, so employment is positively emphasized instead.The loaded term used in the compound is thus used for pragmatic purposes.Portero Muňoz [2011: 145] sees the example as an unusual EFFECT FOR CAUSE metonymy, where the N 2 (gap) "is caused by a lack of" the N 1 (employment).Linguists such as Geeraerts [2002: 471] have argued that there are NNC cases where metaphor and metonymy occur "consecutively […] [or] in parallel [...] [or] interchangeably," which is why Benczes [2009] took another approach in her study.She treated NNCs like belly button as conceptual blends instead, just as Fauconnier & Turner [2002] had earlier studied examples like same-sex marriage as blends.
As this brief survey of the literature reveals, many linguists, psychologists, and other scholars have been concerned with various questions related to nominal compounds and/or euphemisms.While much work has focused on the linguistic properties in nominal compounds, and with the cognitive processes people use to interpret them, we are interested in studying the creation of euphemisms in nominal compound form, including those that may be humorous.To the best of our knowledge, the guided creation of nominal compounds has not often been studied in the past.In the next sections, therefore, we discuss our study and original results.

Study setting and materials
We conducted this study at the 35 th annual conference of the Poetics and Linguistics Association (PALA), an important international organization for scholars doing research in stylistics, literary linguistics, and related fields.The conference took place in July 2015 at the University of Kent in Canterbury, England, and was organized by Prof. Jeremy Scott.At the conference, according to the list of participants, there were 195 scholars from 27 different countries, with roughly equal numbers of Professors and PhD students present.English was the official language of the conference.The data for this study was collected during the presentation by Craig Hamilton on 15 July 2015, entitled "Creative Doublespeak in Euphemisms."Roughly 30 conference attendees were present in the room, and the vast majority were native speakers of English who were at various stages in their careers in different countries.(More precise details of the study population are unavailable because that type of data was not collected in the study.This is because no correlations were to be examined between the results and the group's demographics.)The PowerPoint presentation that was given contained a total of 27 slides, which took about 18 minutes to present, followed by the euphemism creation task (6 minutes), and a brief question and answer session (6 minutes).
All audience members were given two handouts (two pages of A4 paper) at the presentation.The first handout contained a list of 21 attested euphemisms in English, reported first by Portero Muňoz [2011: 155-157] in her list of 67 euphemisms, along with their attested meanings (Table 1).The back page of handout 1 listed the 10 bibliographic sources used then for the presentation.In her study, Portero Muňoz [2011] looked at 67 euphemisms related to the 2008 financial crisis, most of which were attested in The New York Times, with examples ranging from austerity budget to zombie banks, and many other nasty things in between.One of her concerns was seeing how such euphemisms were used as doublespeak.According to Portero Muňoz [2011: 138], euphemisms are usually "less offensive," and they "can save the speaker's face in doublespeak […] typical of governmental, military or corporate institutions."Indeed, in such formal settings, saving face may be akin to trying to make statements that contain a certain amount of "plausible deniability" (Pinker et al [2008]).For instance, after a politician gives a press conference and uses an underhand euphemism, later on an aide might have to clarify the so-called misunderstandings that public outcry has made visible, trying to explain how the politician was misunderstood.
23 The 21 euphemisms in Table 1 were selected from Portero Muňoz's 67 items [2011: 155-157].The 21 were chosen for their alleged transparency value in a decontextualized situation (i.e., isolated phrases in a table, along with their attested definitions).As Table 1 shows, each item was a nominal compound, yielding a total of 42 nouns, with 36 of them used just once.(Portero Muňoz [2011: 148]); or (-) an error (dictionary.cambridge.org) made in a budget

job flexibility
(+) work-life balance when chosen by employee (businessnewsdaily.com)or (-) lack of job security (Pop [2010: 129]) when chosen by the employer

lipstick effect
The more insecure the economy, the more money women spend on beauty products.(Huffington Post, 19.06.2012) 9. membership fees taxes that are ''necessary to maintain the services and infrastructure of the society to which we belong'' (Lakoff [2007: 246]; qtd in Portero Muňoz [2011: 147])

ninja loan
A loan to someone with no income, no job, and no assets (Washington Post, 27.05.2013)

pension overhaul
Usually, pension reduction (-), but overhaul can also be (+) "a process of revision and improvement'" (oed.com)

resource reallocation
Redeploying people and capital (mckinsey.com)

revenue enhancement
Government tax increase (investorwords.com)3).We should note that a few participants did not return their list at the end, and others did not manage to create or define three examples in the time available.Apparently, performance varied amongst participants.Participation was entirely voluntary, and no payment or reward was given for taking part in the study.That factor might have reduced the motivation for some scholars in the room to complete the task.Finally, participants were told verbally that the data we collected would be analyzed later for our study.Nobody who participated openly objected to that; but those who might have done so silently might have decided not to return their papers (handout 2) at the end.2. Thus, these last few findings suggest that the format of Table 2, and how participants read it, might have affected some of their choices.Looking for neighboring words to pair up quickly could have been a factor, too.
Table 4 lists the distribution of the original 36 nouns in the 54 unique examples.Some nouns were selected many times, but two of the 36 were never chosen.The 17 nouns that were used three or more times, on the left side of Table 4, might have seemed salient for participants.The 19 selected two times or fewer, on the right side of Table 4, might have seemed less salient.The fact that 10 terms were selected five or more times is noteworthy, and suggests their salient nature in this study context.For instance, lipstick was the term most often picked (8 times), and the most often used as the N 1 (6 times).It is unclear why, although lipstick is arguably the most concrete noun in Table 2, which is comprised mainly of abstract nouns.The popularity of lipstick might be explained by its association with sensuality, or by its use in other metaphorical compounds, such as lipstick lesbian, which for some language users apparently refers to lesbians with so-called feminine traits.Of course, other factors might explain this result.However, the study was not designed for us to do follow-up interviews with participants.3 and 4, no clearly discernible patterns seem to stand out.Indeed, nouns of high and low frequency in Table 4 appear in various locations in Table 5, which shows how participants used the 17 most frequent nouns.protection, rationalization, and reallocation).One reason for these findings could relate to "nounhood"; some nouns may seem to have had more nounhood qualities than others for participants in the study, so they tended to use them as N 2 position head nouns.This would especially be true for the three nouns in Table 5 that never occurred in the N 1 position: protection, rationalization, and reallocation.Other nouns may have had fewer nounhood qualities in the minds of the participants, so they may have seemed like more likely candidates for the N 1 position.The six remaining nouns (ninja, flexibility, management, pension, reform, and tax) showed no strong preferences for either the N 1 or N 2 positions.Also, career lipstick and lipstick career (Table 3,nos. 4 and 24) are the only examples where the N 1 and N 2 were reversed.Of course, it might be possible to create hundreds of random combinations on a computer, or to categorize them as metonymies (Portero Muňoz [2011]).But the intentional act of creation and definition was our first main focus in this study, and this is what makes it rather different.
Morphological or phonological factors may have also played a role in the selection of nouns and the decision of where to position them.For instance, three nouns that never occurred as the N 1protection, rationalization, and reallocation -end with the -tion bound morpheme and are all three up to six syllables long.Those three nouns are also in the far-right column of Table 2, which might have influenced their selection as N 2 nouns; in Table 2, words appeared to their left, but not to their right.However, counterexamples like enhancement gap (Table 3, no.13) could weaken claims arguing in support of one hypothesis over another, such as assuming that N 1 nouns tend to be shorter than N 2 nouns.We can only speculate here, especially since the examples created by the participants were decontextualized, even if they were defined.But as Ryder [1994: 93] noted, novel NNC interpretation is a "creative problem solving ability" of ours.In her six-stage pragmatic model, listeners can rely on contextual knowledge -or, in its absence, rely on schematic knowledge -or do whatever it takes to make sense of the NNC.This even includes assuming the N 1 -N 2 relationship is one of similarity (Ryder [1994]).Ryder [1994: 95] exaggerated when saying that her model "solves all the problems of previous treatments of compounding" because several studies that came after hers suggest otherwise (e.g.Geeraerts [2002]; Maguire and Cater [2005]; Benczes [2009]; Portero Muňoz [2011]; Jamet [2018], to name just a few).

Discussion
Our first research question was as follows: "How creative can people be when asked to produce a few novel euphemisms in a very short time with the same starting conditions (a list of 36 nouns)?"As we saw in Table 3, up to 14 of the 54 items might not seem creative.It is either because they are not that new, or because of their initially close proximity in Table 2.This leaves us with 40 apparently original examples, a number which suggests a certain degree of creativity, albeit constrained by the 36 nouns listed in Table 2 and derived from Portero Muňoz's examples (Table 1).Moreover, no true neologisms were created in the study task since all nouns we provided were relatively common in current English usage, and no participants changed their form to a word from the same word family (e.g., turning tax into taxation).Thus, original neologisms are not measure of creativity here.What is more, most items in Table 3 might be found on the internet, with various rates of frequency, so it seems that no entirely new NNC was created either.That said, in gap job (Table 3, no.18), one instance where gap appears as the N 1 , its definition -"students who work before university, they work for a year, then go back to university, then do a job etc." -suggests that it is inspired by gap year, the well-known British collocation.However, Benczes [2009: 50] claims that one reason we use euphemisms is "to come up with novel expressions with a rich mental imagery," and some examples in Table 3 do seem more vivid than others, such as currency ninja compared to workforce flexibility.
Be this as it may, the amount of creativity or originality in an example might be seen in the similarity or dissimilarity of the nouns in the nominal compound.Because words like income and tax may come from the same semantic field, they may appear to be less creative in combination than words that are less similar, such as lipstick and gap (Table 3, no.25), which was defined as "gender inequality in pay."So the importance of N 1 -N 2 similarity that Ryder [1994] pointed out may relate to impressions of creativity.Indeed, pre-existing relations between nouns might affect how creative or humorous certain examples seem.What is more, the "camouflage" hypothesis of McGlone et al [2006] may also explain why common collocations like income tax (Table 3, no.22), may strike us as less creative than vivid examples like lipstick gap (Table 3, no.25).The nearly "invisible" example might go unnoticed in a conversation, while the creative and lively one would get noticed.
34 Granted, not all examples in Table 3 may look like euphemisms, but participants did seem able to produce some interesting examples which they intentionally defined as having euphemistic meanings.This intention is key to our understanding of what a euphemism is here.For instance, budget ninja (Table 3, no.2) was defined as a compliment -"someone good at budgeting" -while income tax (Table 3, no.22), a common collocation, was defined rather neutrally as a "social contribution to society." In fact, while creativity might not always be seen in some nominal compounds themselves, it can nevertheless be seen in the definitions.For instance, resource management (Table 3, no.45) is not an entirely original collocation, but its novel definition is: "no raises for you."Other examples seem both creative and humorous.
For instance, lipstick management ("board of managers consisting of females only"), lipstick pension ("savings destined for keeping high maintenance lifestyle intact"), and revenue rationalization ("kids taking money from parents") seem to reflect creativity in their definitions (Table 3,nos. 26,27,and 48).The term lipstick management, however, might also be used as a sexist remark if men use the term to complain about their women managers, perhaps because of the connotation of superficiality or frivolity that lipstick may have in this case.In sum, the fact that 54 of the 61 compounds were coined at least once in the task, and that they all have rather creative definitions, suggests that participants demonstrated a certain degree of creativity here.As Portero Muňoz [2011: 139] claimed, the "semantic obscurity that characterizes noun-noun sequences makes them likely candidates for euphemism creation," and our results appear to support this observation.
35 Our second research question was as follows: "Is it true that the more neutral the meaning of both N 1 and N 2 , the more likely it is for the NNC to be used in doublespeak?"Table 3 shows that most examples that participants created seem to relate to business.This is because the original examples from Portero Muňoz [2011] in Table 1, and thus the input nouns in Table 2, are mainly about business.Many of the nouns in Table 2 are abstract, and vary in their semantic prosody.For instance, enhancement or protection or stimulus seem positive, while fee or tax or rationalization seem negative.Neutral terms might include currency or resource or package.Whether a noun in isolation is positive, neutral, or negative is one thing, but its significance can change once it enters a nominal compound (Ryder [1994]; Geeraerts [2002]; Maguire & Cater [2005]; Benczes [2009];Portero Muňoz [2011]; Jamet [2018]).For example, flexibility loophole -"your flexible schedule can be made less flexible" (Table 3, no.15) -is an example where both nouns, which may seem neutral in meaning, can become negative in meaning when used in combination to form a euphemism used for the purposes of doublespeak.In other words, the N 1 and N 2 may influence the meaning of each other, while the meaning of the unit as a whole can behave differently in usage as well.As Table 3 On Euphemisms, Linguistic Creativity, and Humor Lexis, 17 | 2021 reveals, there are just four items which seem neutral in connotation: income tax, lipstick pension, shelter pension, and user rationalization.In contrast, most examples (41 out of 54) are negative in connotation; thus, they could be called "underhand euphemisms" that might be used as doublespeak.
Moreover, there are nine ones which seem positive based on their definitions in Table 3: budget ninja, currency reallocation, employment shelter, fee shelter, gap job, lipstick protection, loan protection, loophole reform, and revenue loophole.These might be uplifting euphemisms, meant to give something simple a more flattering name in technical jargon.As most of the N 2 nouns in these positive examples already have positive connotations, it suggests that "yes" is the answer to our second question.
Regarding humor and euphemisms, Colston [2015] uses the term "pragmatic effect" to refer to effects felt by listeners and speakers in response to figurative language.
Colston's taxonomy of pragmatic effects [2015: 66-70] include six general pragmatic effects, which range from "ingratiation" to "efficiency," and thirteen specific pragmatic effects (linked to figures), which range from "expressing negativity" to "tension reduction" (Colston [2015: 71-84]).Just as a figure like metaphor can create several different pragmatic effects, one specific pragmatic effect, like "highlighting discrepancies," can be caused by various figures, such as hyperbole, or by irony (Colston [2015: 74]).The cause and effect relationships in figures of speech are rich yet sometimes opaque, and boundaries between them are sometimes fuzzy.Having said that, while Colston [2015: 76] admits that "the relationship between humor and figurative language is one of enormous complexity," this is perhaps because "the indirectness […] of all figurative language […] could trigger humor" (Colston [2015: 75], italics in original).Examples of uplifting euphemisms, such as intellectually challenged (dumb), vertically challenged (short), or horizontally challenged (fat), may make audiences laugh when hearing them for the first time, perhaps because audiences recognize them as blatantly designed to avoid offence or save face.At this point, they could become ludic euphemisms.However, dysphemisms in informal settings might also be humorous, since their frankness could surprise and amuse audiences.
As for humor and euphemisms in our study, there was some notable laughter during the presentation in 2015, and when participants saw the examples in Table 1, or wrote down their examples and defined them afterwards.However, this pragmatic effect was not measured in an objectively scientific manner.And the situation can be more complicated than seems at first sight.In some cases, as Colston [2015: 153] notes: Humor occurs without laughter.Laughter occurs without humor.Correspondences between humor and laughter when they do co-occur are very complex -the causal direction go either way, other causal factors can intervene between humor and laughter (and vice versa), and external causes can affect one but not the other or both.
Informally, however, the generally positive reactions to the presentation in 2015 suggest that at least some euphemisms in Tables 1 and 3 were indeed deemed to be humorous.
Indeed, it is not difficult to imagine situations where some examples in Table 3 could be ludic euphemisms.For instance, lipstick management could be used for a humorous pragmatic effect, if said by a man amongst men to make them laugh or smile.The euphemism's semantic value could vary from neutral to negative to humorous, depending on the context of usage, and the intention behind the usage, as would be expected for X-phemisms, or what Crespo-Fernández ([2015: 46)] calls "quasieuphemisms".For Pinker et al [2008: 833], human communication is "a mixture of cooperation and conflict," and examples like lipstick management may reflect this, depending on how the term is used, where, when, why, and by whom.For instance, a person might use dysphemisms with one's superiors in a company, even though Cowen [2010: 5] remarked that "[r]eal 'straight talk' very often is not compatible with authority, as it breeds conflict."But the dysphemisms could been seen as humorous or ludic euphemisms when the worker retells the story to friends later on.In the first setting, the company bosses might not have found the dysphemisms funny, but the friends in the second setting just might.
40 Such judgments relate to interpretation.In dramatic irony, the audience has information that certain characters on stage or in a story do not.Likewise, overhearers who recognize a euphemism as flattering for the intended hearer, may be amused by it, even if the hearer at first does not recognize it as intended to be funny too.Indeed, a listener's willing "resistance" to humor may occur sometimes (Colston [2015: 223]).Yet, as Simpson [2004: 45]  41 Our preliminary findings suggest that some euphemisms in this study might also be used for humor.For Colston [2015: 75], humor is a specific pragmatic effect that could "arise as a consequence of some other effect," such as ingratiation, one of the general pragmatic effects.We can make indirect compliments by using figurative language in order to flatter listeners (Colston [2015: 67]), and their positive response in turn can have a positive effect on us.While we can use euphemisms to avoid offending someone, we may also use them to get on someone's good side.A ludic euphemism could thus achieve the pragmatic effect of humor when the result is effective.What is more, when we recognize an underhand euphemism as a euphemism, we might be likely to find it humorous, and thus a ludic euphemism.This might complicate the camouflage hypothesis, which holds that the harder it is to recognize a euphemism as a euphemism, the more effective it is.

Conclusion
In this study, we presented euphemisms and their definitions created by participants in England during an academic conference.This project is clearly more qualitative than quantitative, and was loosely designed as such from the start.With our data, no robust correlations can be made concerning the euphemisms and their creators' demographics.Also, in the age of big data, this project is small.It presents authentic results, yet limited in number.
We set out to answer one question about creativity in euphemisms, and we found evidence of some creativity in the euphemisms that our participants created.But some examples (e.g.lipstick tax) seemed more creative than others (e.g.income protection).We saw more creativity, though, when participants defined the euphemisms they had coined.While some examples they produced might not seem original, their definitions often were.For instance, while income protection (Table 3, no.20) has been used before (Holder [2007: 225]), its previous definition differs from the novel one given here by a participant in this study, "firing other people to make sure you retain the same rate of pay."Moreover, even when the same combination of nouns was created seven times as a euphemism, it was defined differently each time, yielding 61 definitions for 54 examples (Table 3).As Colston ([2015: 168]) states, "Creativity arises and interacts with figurativeness and pragmatics effects," and our findings offer some evidence for this kind of linguistic creativity.
Our second question was about doublespeak and the nature of the nouns used in euphemisms that are nominal compounds.Most of the examples seemed likely candidates for doublespeak as many of them were "underhand euphemisms," to use Burridge's term [2012].How positive or negative a single noun is might not always be obvious when it is decontextualized.Holder [2007], for instance, even lists single nouns as euphemisms in his dictionary.Meanings can change when nouns are combined, and when the nominal compounds are used as euphemisms.The 54 euphemisms produced for this study need to be explored in more depth in various corpora.Data on usage could tell us more about their meanings, their semantic prosodies, and even their evolution through time.It would also tell us more about their use as doublespeak, why some nouns were more popular than others (Table 4), or why some nouns tended to occur more as N 1 than N 2 (Table 5).
Finally, we also discussed how euphemisms could be humorous.Humor is an important pragmatic effect of what Burridge ([2012: 71]) calls "ludic euphemisms."Our results show that euphemisms might fit into several categories at once.In one setting, an underhand euphemism may be used as doublespeak, but it could become funny, and thus a ludic euphemism, in another situation.Again, more corpus data might tell us for what purposes and in which contexts people use some of the examples presented here.However, the small sample of specifically created euphemisms we have reported here does enable us to say that nominal compounds may be creative, euphemistic, and even humorous.But as usual, more research remains to be done.To put it gently, this is not the last journal article on euphemisms.

Table 1 :
List of business euphemisms and their definitions The input items were the 36 unique nouns in Table2, which was also projected at the same time on the final presentation slide.This was so participants could see them before receiving the second handout, which took a minute or so to distribute.No nominal compounds in Table1appear side-by-side in Table 2. Items in Table 2 were simply listed in alphabetical order, like the examples in Table 1.Only tax gap (Table 1, no. 16) contains words that appear in the same row of Table 2 (penultimate row, 11).
24 The 36 nouns used once in Table1were reproduced in Table2, and provided to audience members on the second handout (one A4 page, single sided).On the second handout, we gave participants this simple instruction at the top of the page: "Using items from the list below, please create 3 original nominal compounds (N 1 +N 2 ) that could be used as euphemisms and briefly define them."

Table 2 .
Presentation of the 36 nouns Based on the 36 nouns in Table 2, our participants created 61 nominal compounds and defined them, with 54 unique examples as the main result (Table

Table 3 :
Nominal compound euphemisms created in the taskAs Table3shows, two different participants made the same nominal compound five times(nos.24,29,35,37,and 46), but gave them different meanings.Meanwhile, one compound (no.21) was created by three different participants, but also given different definitions each time.In the end, 54 items had single definitions, while seven did not.Among the 61 definitions given for the 54 examples in Table3, only a small number can be said arguably to have a positive or even neutral definition(nos.12, 18, 22, 26, and 37, 29. lipstick tax(1) a requirement to kiss a person; (2) a gender based tax 30.loanprotectionsense of security which might release of the (felt) burden of loan definition 2).In fact, some examples seem to be instances of dishonest doublespeak, or underhand euphemisms(Burridge [2012: 68]), such as resource management (no.45),meaning"noraises for you."Meanwhile,othersseem to be instances of uplifting euphemisms(Burridge [2012: 69]), such as the apparently bureaucratic revenue rationalization (no.48), defined by the participant as "kids taking money from parents."Of course, what might seem like an uplifting euphemism for the kids might be seen as an underhanded euphemism by their parents.When we look more closely at Table3, several things need to be pointed out.First, income tax (no.22) is not new, nor is its definition highly original: "social contribution to society."Second,incomeprotection(no.20)alreadyappeared on Table 1 (no.6).So this example is also not new, although its definition is: "firing other people to make sure you retain the same rate of pay."Third,careerlipstick,flexibilityoversight,loopholereform, and ninja shelter (nos.4,16,32, and 37) already appear as nouns side-by-side in Table2.Fourth, career lipstick, lipstick career, pension adjustment, resource management, and workforce flexibility(nos.4,24,41,45,and54)combine two nouns that come from the same rows in Table2.Fifth, management revenue, ninja revenue, pension protection, and revenue shelter(nos.34,36, 42, and 49)are comprised of nouns from cells that are either horizontally or diagonally adjacent in Table

Table 4 :
Noun frequency in the task

Table 5 :
Positions of the most frequent nouns notes, humor often results from incongruity, and one aspect of what he calls "the principle of incongruity" can involve "any situation where there is a mismatch between what someone says and what they mean."WhileSimpson'sdiscussionfocusesmainly on dramatic dialogues from absurd plays, his insight can easily apply to euphemism.Saying you work in employment management, when it actually means you work "in the department in charge of firing people" (Table3, no.11), could turn this underhand euphemism into a ludic euphemism if the doublespeak attempt fails, and produces instead a humorous reaction.As Colston  recognized [2015: 64], "The juxtaposition of […] positive commentary and negative event is […] an incongruency that, at least in some humor theories, is a necessary condition of humor."In other words, employment management may be pure Orwellian doublespeak in one context, yet humorous in another, when the apparently positive euphemism is noticed to clash with the negative reality it hides.Finally, as the work of Simpson reveals, dialogues in modern or contemporary plays might be an interesting source to explore to find more examples, and to see how readers, audiences, or even other characters on stage react to euphemisms.