For the Love of Prepositions Part 13 — For it is sweet…

A recent translation forum debate regarding the correct preposition to use in the names of government departments reached a consensus that, while ‘of’ is the most appropriate preposition for the more prestigious ministries, lesser government agencies with longer more descriptive names tend to be less resistant to the use of ‘for’. ‘Ministry of’ is thus the proper nomenclature for Health, Education or Defense, but ‘Department for the Environment, Housing and Planning sounds OK too.

While few countries around the world have graced their major offices of state with the title of ‘Ministry for’, the phraseology is certainly very common in translation and this suggests that some underlying change is in fact afoot. In the case of government agencies that opt to call themselves departments rather than ministries, the changeover from ‘of’ to ‘for’ is already quite advanced.

In 1992, in the UK, for example, the one-time Ministry and then Department of Education (in existence since the Second World War) was transformed into the Department for Education and, despite various infelicitous name changes since then, the ‘for’ has remained firmly in place.

A similar pattern emerges with other UK government departments. The Department of Transport swapped its ‘of’ for ‘for’ in 1997, and various other newfangled divisions of Her Majesty’s Government have been baptized likewise in recent years.

The purpose of this shift to ‘for’ is clear. ‘For’ connotes a caring agency rather than a grim bureaucracy and suggests the friendly encouragement of self-help rather than the doling out of benefits. For the same reason, the pompous- and—since George Orwell published 1984—somewhat sinister-sounding ‘ministry’ is being pushed out in favor of ‘department’ or ‘office’, both of which are words and concepts that voters will be familiar with from their own workplaces.

Even Orwell of course named his dystopian ministries the Ministry of Truth, the Ministry of Peace, the Ministry of Love. ‘For,’ in this context, would in fact sound even creepier. Perhaps we have good reason to be wary of ‘for’.

A quick Google search for “Ministry for Truth” shows that the name of the authoritarian institution in 1984 is in fact frequently misquoted this way. More interestingly still, this usually occurs in contexts that are themselves clearly using the term for propagandist purposes.

A Sky News podcast likens Antifa to Orwell’s Ministry for Truth (sic). The Daily Telegraph warns us that Orwell warned us of a coming ‘Ministry for Truth’. And, in a peculiar twist, the Guardian newspaper misquotes the Sun quoting Orwell’s dystopian novel correctly when accusing David Cameron’s government of operating a ‘Ministry of Truth’.

A left-wing blogger freely moves back and forth from ‘Ministry of Truth’ to ‘Ministry for Truth’ in a single post attacking Teresa May, and Time Out, in a review of a lefty gig, remarks that the University of London’s Senate House provided the architectural inspiration for Orwell’s ‘Ministry for Truth’.

Both extremes of the political spectrum would thus seem to agree that “Ministry for…” sounds far creepier than “Ministry of…” and yet this is precisely the terminology that is becoming fashionable and finding increasing favor in government circles these days.

A recently published science fiction novel about environmental catastrophe entitled The Ministry for the Future has won plaudits from Barack Obama and other liberal politicians. The imaginary international organization that gives the book its title is described as being ‘“charged with defending all living creatures present and future who cannot speak for themselves” and is clearly intended to inspire our praise.

Herein, however, lies the problem with ‘for’. An organization that speaks or acts for groups of individuals who would otherwise be deprived of a voice can be seen as liberating in so far as it advocates for the rights of those who go unrepresented in modern democracies, but it could also be viewed as sinister in so far as it presumptuously arrogates to itself the right to speak for groups (such as animals or the unborn) who are a priori incapable of speech or opinion and, by extension, for other supposedly ‘silent majorities’ who do in fact have their own voice.

As the ideological diversity of the misquotations from Orwell cited above suggests, no one single party or political persuasion has a monopoly on this tendency. Activists on the left claim to speak for the workers or the planet, while those on the right stand up for the nation and the war dead.

‘For it is sweet and honorable to die for one’s country’, a long-dead poet once pronounced on behalf of heroes who had long since lost their right to voice their own point of view.

‘For’ is a preposition that has a long history of sweetening such subterfuge.


The Politics of Stress

[In the final week’s runup to the US presidential elections, I publish two posts on language related to this event. This is the first of them.]

The Politics of Stress

The prosody of Present-Day English is dominated by stress. This means that certain syllables in a word are given greater or lesser degrees of emphasis. A stressed syllable is usually louder than an unstressed one and may also differ in pitch.

Words of two syllables in English are usually stressed on the first (DUM-di), as in ‘MOther’ and ‘LANGuage’. Sometimes, however, the stress falls on the last syllable. This can be used, in some instances, to distinguish nouns from verbs. Note, for example, how the difference in stress in the word pairs ‘REbel’ : ‘reBEL,’ ‘UPset’ : ‘upSET’ distinguishes the noun (the first of the pair) from the verb (the second), even though these are indistinguishable in spelling.     

The most common stress pattern for a three-syllable noun in English is DUM-di-di. In this, English follows the practice of its Germanic neighbors and forerunners.

Deviations from this norm, therefore, frequently indicate a word of foreign non-Germanic origin.

A stress on the final syllable, for example, often occurs in the names of exotic animals, such as kangaroos, or exotic dishes such as vindaloo. Likewise, a stress on the penultimate syllable marks a word or name as foreign in origin: potatoes, tomatoes, and volcanoes are not native to English soil.

Given names tend to follow the same general pattern. The stress falls on the antepenultimate syllable in traditional English or invented names, but can fall elsewhere in names of foreign origin or names on which the giver intends to confer an exotic feel. The latter, for some reason, tend to be more common among women and girls.

Thus, more traditional, anglicized or obviously Germanic girls’ names tend to be stressed on the first syllable in three-syllable names: Márgaret, Ísabel, Cátherine, Álison, Híllary, Émily, Hárriet, Stéphanie, Béthany, Mádison and so forth all obey the rule. [Stressed syllable marked artificially by an acute accent] The same applies to names of more than three syllables, such as Elízabeth and Victória. Invented names such as Pámela, Jénnifer and Jéssica, irrevocably fall into step with the usual pattern.

Names that diverge from this tend to have a deliberately foreign or classical sounding feel to them. Historically, such names have often been preferred by aristocratic elites—as Diána and Camílla both attest, but also among post-Civil Rights Movement African Americans keen to disassociate themselves from names given them by white slave owners. In recent years, this fashion has spread more widely to all sectors of society.

There is therefore good evidence to suggest that modern English has a strong preference for a stress on the antepenultimate syllable of polysyllabic words (including names) and that stress placed elsewhere is associated with loan words or foreign names.

As a corollary of this, English speakers will tend to mispronounce unfamiliar foreign words by moving the stress from penultimate to antepenultimate position, especially if the penultimate syllable is followed by a single consonant. When this ‘error’ becomes the norm, it suggests that the name, concept or word borrowed from abroad have been fully assimilated into the host language. Márgaret is fully Anglicized; there is no hint that the original French name was stressed on the last syllable Marguerite. No-one would call Margaret Thatcher, Marguerite Thatcher, unless they were making some heavy-handed point about the un-Englishness of her ways.

Speakers, may also, however, do the opposite and hypercorrect, placing the stressed syllable on the penult in foreign words even when it is incorrect to do so, in order to make them sound more foreign. This may, as in the recent case of mispronunciation of US vice presidential candidate Kamala Harris’s first name by Republican senators and the current US President, be done deliberately. In this case, the error clearly serves the explicitly racist purpose of making the individual seem more alien than she in fact is. In this context, such behavior is a form of linguistic exclusion or shunning.

Kamala is a name derived from the Ancient Indian language, Sanskrit. It means ‘lotus flower’. The name is stressed on the first syllable in the original Sanskrit and thus follows exactly the same stress pattern as that of the vast majority of two-syllable English words. Furthermore, in view of Ms. Harris’s status, the correct pronunciation of her first name is well known in political circles. There is, therefore, no reason other than racism to pronounce the senator’s name otherwise.

Modal Verbs in the News

Grammar seldom makes the news. In fact nothing seems to make the news much these days except for Brexit and Trump.

I made a conscious decision early last year NOT to write about the US president on this blog, even though politics is one of its main thematic axes. Since Trump seems to thrive, like some extremophile strain of virulent bacterium, even on negative publicity, I feel it is the best policy not to give him any publicity at all.

I make an exception, however, where language is concerned; and, since I have written extensively on modal verbs in past posts, I feel obliged to comment when one of them makes the headlines, as has occurred recently, even though it came out of the mouth of Mr. T.

At the Helsinki summit press conference, when asked a question regarding Russian tampering in the 2016 US elections, Mr. T uttered the following words: “I don’t see any reason why it would be [Russia]”

After much outcry over this statement, not least among members of his own party, the very next day the US president retracted his words, claiming that he had ‘misspoken’ ‘would’ instead of ‘wouldn’t’ and thereby, in an unimpressive display of clumsy prestidigitation, completely inverted the apparent original meaning of the phrase.

Leaving aside questions regarding the genuineness of this self-correction or the nature of Mr. T’s true feelings towards Vladimir Putin and the Russian Federation, I think it is worthwhile taking the president at his word on this occasion and examining the ‘mistake’ in purely linguistic terms.

To be fair, Trump himself, in the course of his retraction, explained the error in precisely this way. “It’s a sort of double negative,” he added (off-script) after reading out the corrected version of the original statement.

This provides a glimpse of insight into what really goes on in Mr. T’s mind, whether his intentions were duplicitous or not.

Trump is old enough to have been schooled quite sternly in a kind of prescriptive grammar, which, like much of what Mr. T believes in, is rightly considered politically incorrect and scientifically wrong these days. One such ‘rule’ that would have been beaten into him (perhaps literally) as a young student at a posh military school in the 1950s concerns avoidance of a so-called ‘double negative’.

The reason for this, of course, is that phrases of the kind “I ain’t done nothing” tend to be stereotypical markers of low social status and, in the US, non-white ethnicity.

The truth is that, from a descriptivist point of view, standard middle-class white English, on both sides of the Atlantic, is a language in which multiple negative markers are interpreted ‘logically’ rather than emphatically. Thus, “I haven’t done nothing’ is interpreted to mean “I have done something.” The two negatives cancel each other out just as they do in mathematics. I should add that, in this example, the double negative adds a sort of whining apologetic tone to a neutral double-positive statement and sounds sort of ‘off’.

But this logical interpretation of double negatives is by no means a linguistic universal nor even true for the wide range of brands of English now spoken around the world. Many languages, including my own second language, Portuguese, use multiple negative markers primarily to emphasize the negativity. Logic doesn’t come into it. No means no. And three no’s mean “No! No! No!”

Even within the English family of dialects, this emphatic use of multiple negatives is accepted in certain contexts and is common in cultures whose language use is colored by an ancestral substrate.

Going back to Trump’s sentence, however, we should note that it is not in fact a double negative in the sense outlined above. The two negative markers clearly occur in separate clauses, one embedded in the other in the manner of indirect speech.

In such cases, the logical interpretation of negatives applies in all cultures and all languages. There is a world of difference between “I never done nothing,” which is clearly emphatic and “I never said I never done nothing”, in which the two negative clauses cancel one another out.

So, there is nothing grammatically wrong with what Mr T. originally said and no reason for a mea culpa or correction on these grounds.

This is what linguists and language teachers call hyper-correction. The speaker or writer makes a mistake because they are inappropriately or over-exactly applying a ‘rule’ learnt by rote.

However, as is so often the case when we start to analyze discourse linguistically, there is something else going on here. And this concerns not the old-fangled notion of double negatives but a very modern shift in the use of the modal verb ‘would’.

There are no prescriptive sociolinguistically-determined rules regarding the use of ‘would’.

Supposedly descriptive grammar text books aimed at non-native speakers have traditionally typified ‘would’ as a ‘translation’ of the Latinate conditional. It usually presented in the context of sentences containing ‘if-clauses,’ in cases where the outcome is unlikely or counterfactual, although the modal verb is obviously not confined to this structure and context.

In fact, as Michael Lewis has pointed out, this traditional exemplification and implicit characterization of the use of ‘would’ is misleading to say the least. Numerous language students of mine have noted that ‘would’ does not occur most frequently in English in relation to an conditional, nor is ‘would’ the most frequent modal verb used in conditional sentences.

Matters are further confounded by the fact that there is a strong tendency, even in erudite discourse, and especially in the United States, to use ‘would’ in both halves of a sentence involving an ‘if-clause’.

Interestingly, this grammatical deviation from the ‘norm’ tends to go unnoticed and is certainly not subject to the kind of prescriptivist opprobrium associated with double negatives. The veteran CNN journalist, Wolf Blitzer, does it all the time.

There are in fact some good reasons for this structural development. The double use of ‘would’ sets up a nice parallelism that adds clarity, especially in situations where clarity is especially important, such as those involving reporting or reading the news.

There is, however, another more negative tendency to use ‘would’ very generally and vaguely, merely to distance or exempt the writer or speaker from responsibility for the statement uttered. It is habitually used by journalists and politicians and is even spreading at an alarming rate to academic discourse as well. In Latin languages, the equivalent use of a conditional has become standard in all fields of discourse to the extent that it is practically obligatory in any public statement of an opinion and even in statements of facts.

I get why people do this. It sounds modest and polite and acknowledges differences of opinion and the need for scientific doubt. However, it also threatens to cast a veil of suspicion over any statement uttered (however self-evident or well-documented) and, at the same time, provides a convenient safety-valve of potential retraction for those who wish to benefit from the dissemination of patent untruths.

It lays the linguistic groundwork for a world in which all knowledge is suspect, all facts questionable, and in which all promises, commitments and principles can easily be turned on a dime.

That may be the kind of world in which the likes of Mr. T prefer to operate, but it is definitely not the kind of gloomy unpredictable dark age most people in this world would ever wish to live in.


Ergativity in Trump White House Discourse

In a recent press conference on the resignation of Michael Flynn, Trump’s spokesperson, Sean Spicer, used the phrase “There is nothing that would conclude me that…”

There is something odd about the way this is formulated.

‘Conclude’ is a verb, like many in the English language, that hovers between transitive and intransitive status, but has a slightly different meaning in each case.

Transitive ‘conclude’ is more or less synonymous with ‘finish’, with extra emphasis on the finality of the process, or a ceremonial flourish. The two meanings may, of course, as is often the case, overlap.

Intransitive ‘conclude’ is more abstract. It can mean ‘draw a conclusion’ if the agent is human, or simply ‘end’ if the subject is inanimate, again, with slightly more emphasis than the near synonyms. Like other verbs of saying or thinking, ‘conclude’ can also be followed by a ‘that’ clause. Arguably, this should be construed as a transitive usage.

Linguists divide languages broadly into those that are ergative-absolutive (ergative for short) and those that are nominative-accusative (accusative for short), although this is by no means a rigid distinction. Most languages contain elements of both.

In an accusative language, the subject/agent of both an intransitive and a transitive verb share the same form or position in the sentence, while the object of a transitive verb has a distinct form or position. In ergative languages, the subject or agent of a transitive verb has a distinct form, while the subject of an intransitive verb and the object of a transitive one share the same properties.

It is generally the case that, in accusative languages, the object form is viewed as deviant, while the subject form deviates little from the root form. In ergative languages, the opposite tends to be the case: the agent of a transitive verb has a special form, while the object/ergative subject is akin to the root.

English is essentially an accusative language, although this has been eroded over time, as accusative endings, with the exception of some pronouns, have disappeared, and word order has become the main determinant of the relation between the subject and object of a verb.

Many linguists however have pointed out that there are many ergative features of English, including so-called ergative verbs, and that these may be growing in number in recent years. English may thus be going through a transition from a primarily accusative to a semi-ergative language.

Phrases such as

  1. The door closes.
  2. This car drives well.

suggest that there may be a shift towards using the pre-verbal position to indicate the object of an intransitively-construed transitive motion or action.

This, of course, is by no means a fully ergative form, but suggests that English may be drifting in that direction.

Remember that, in accusative languages, the object is regarded as deviant or subordinate (that is what declension means), while in ergative languages, it is the agent of a transitive verb that is viewed as a deviation from the norm of an otherwise static peaceful impersonal world. It is perhaps no coincidence that most ergative languages are very ancient (e.g. Sumerian) or have very ancient roots (e.g. Basque, Australian first nations languages). They stem from an age in which human agency played a much less important role than the timeless impersonal processes of the natural world.

Let us go back to Sean Spicer and try to unpack the underlying unconscious meaning of his words. I deliberately leave the content of the ‘that’ clause blank; it matters little to the argument I am presenting here, which concerns more the way supposedly factual statements are now being framed and is relevant to the current debate about ‘alternative facts’ and ‘post-truth’.

First of all (although this is not the most important point), note how clipped the phrase is. Grammatical purists, who turn their noses up at Twitter and TV, would argue that it should read “There is nothing that would lead me to the conclusion that…” or more succinct “Nothing would lead me to the conclusion that…”

But neither of these formulations works on Twitter or TV. The first is two wordy; an impatient TV sound-bite listener or keystroke-stingy online reader is already bored by the time we arrive at the second ‘that’. The second sounds negative from the start, with nothing as the subject of the verb. “There is” tempers “nothing”, establishing a more positive tone.

Next note the deviant use of the ‘indirect object’ “conclude me”. This is an extension of ‘tells me’ etc. but also, perhaps more importantly, marks a clear contrast with the bald absolute “I conclude…” Spicer of course is well aware that his function is that of a ‘spokesperson.’ He is supposed to avoid the first person singular, but there is no reason not to use “the president” or “the administration” as the agent of the sentence.

The formulation thus pushes away any kind of agency or responsibility. The conclusion ergatively drives itself.

Such a formulation allows for a lot of wiggle room. Challenged as to the veracity of the statement, Spicer can fall back on a “But I didn’t say that….” Nor does he have to appeal to some abstract agency “the American people,” “we”, “God” or even his boss.

We see the same phenomenon in Trump’s thumpingly repeated slogan “Make America great again!”

This could be construed as an imperative exhortation, but makes little sense if read that way. Neither does it fill the zero subject place with an individual or collective subject. Despite Trump’s obvious narcissistic tendencies, his discourse is strewn with this kind of ergative-minded abstract subterfuge. You could be forgiven for thinking that his favorite pronoun is “I”. In fact he more often uses ‘it’, ‘this’, ‘that’, or, given his penchant for the brevity of Twitter, simply zero. The ingenuousness and the sense of responsibility conveyed by Obama’s “Yes we can,” is instantly undermined.

Ergative language is fundamentally alienating and fetishizing in the Marxist sense. The door which closes is either closed by someone, blown by the wind, or programmed to do so. A car drives well because thousands of workers and designers and engineers have labored to manufacture it.

There is no invisible hand, only the hands of thousands and millions of agents. Trump with his small lily-white hands that have never fixed bricks in mortar themselves can perhaps be forgiven for not understanding that.

The Genuine Article

As readers of my previous linguistics posts will have noted, I am a great believer in the potential of statistical and computational linguistics. Such valuable data, however, need to be treated carefully and set in an appropriately designed and historically-sensitive theoretical and explanatory context.

Many languages do not have articles, although most of these do have some kind of emphatic particle or demonstrative pronoun that can, if need be, serve this role. Even within the Indo-European family of languages, there is much variation—Latin and Sanskrit have none, Ancient Greek one, neo-Latin and neo-Germanic languages a full range of three. Russian has none.

In the longer-term history of modern Western European languages, therefore, it would be reasonable to suppose that there has been a general trend towards greater use of articles, spurred, presumably, at least in part, by a greater interest in distinguishing between degrees of definiteness.

This, however, begs a crucial question. What do we mean by ‘definiteness?’ How do we ‘define definiteness,’ in a multilingual context in which we must admit that many languages lack the means to express it? The more we examine it, the concept of definiteness becomes ever more slippery, ambiguous and hard to define.

So-called definite and indefinite articles perform a wide range of functions in present-day English and these can certainly not be exhaustively described using a theory based on a binary distinction between definiteness and its opposite.

It is also important to note the existence of a ‘zero article’ in English, meaning that the distinction is non-binary—a choice always being available between at least three options. Zero-articles (like other zero-elements in linguistics) are harder to track statistically or computationally, although they clearly bear much semantic weight.

If, as various recent studies have suggested, the use of the definite article ‘the’ in English has been on the decline in the more recent history of the language, we must entertain the possibility that this loss may have flowed equally in the direction of the ‘indefinite’ and the ‘zero’ article or elsewhere.

It is therefore worthwhile not only counting words, but also examining how they are used in particular historical contexts and what the use of an alternative phraseology might entail in semantic and hence ideological terms.

I take as my starting point two texts that were recently juxtaposed in the Guardian newspaper as exemplifying this supposed trend towards decreasing use of the definite article. The first comes from the opening to George Washington’s State of the Union address in 1790, the second from that of Barack Obama in 2014.

Here first is the George Washington text. To provide a more comprehensive count, I have flagged zero articles as [0] and highlighted articles and other relevant words.

“I embrace with [0] great satisfaction the opportunity which now presents itself of congratulating you on the present favourable prospects of our public affairs. The recent accession of the important state of North Carolina to the Constitution of the United States … the rising credit and [0] respectability of our country, the general and [0] increasing good will toward the government of the Union, and the concord, [0] peace, and [0] plenty with which we are blessed are [0] circumstances auspicious in an eminent degree to our national prosperity.” George Washington 1790

The count here, including the zero article is:

Zero = 6

Definite = 11

Indefinite = 1

The Obama text runs as follows:

“Today in America, a teacher spent [0] extra time with a student who needed it, and did her part to lift America’s graduation rate to its highest level in more than three decades. An entrepreneur flipped on the lights in her tech startup, and did her part to add to the more than eight million new jobs our businesses have created over the past four years. An autoworker fine-tuned some of the best, most fuel-efficient cars in the world, and did his part to help America wean itself off [0] foreign oil.” Barack Obama 2014.

The count is:

Zero = 2

Definite = 5

Indefinite = 4

However, this fails to take into account the use of genitives in apostrophe s and possessive pronouns, which serve as proxy definite articles in English, making the use of the definite article marker redundant.

Washington’s text includes three possessive pronouns, all of them, interestingly, second person plural, and no apostrophe s’s, bringing the definite article count up to 14. Obama’s text contains six possessive pronouns and one apostrophe s, bringing his definite article count up to 12. We could add to this the fact that Obama prefers to use the less formal and technically inaccurate ‘America’ rather than ‘the United States.’ Since non-descriptive proper names count as proxies for definiteness in modern English, these too should arguably be included in the tally of definite articles or their proxies. This lifts Obama’s definite article or proxies count to 14, equaling that of George Washington.

The shift therefore is not so much from one kind of article to another, but one involving the increasing use of genitives, possessive pronouns and proper names. Neither does this indicate a decline in either definiteness or formality (the contours of which shift randomly over time and tend to even out), but rather a growing concern with individuality and specificity.

The more interesting and perhaps statistically significant change, may, if we count the articles the way I propose, lie in the use of the indefinite article. Washington uses it once and in a peculiarly old-fashioned sounding manner that has a distinctly definite flavor to it. Obama uses it four times, on each occasion both to generalize and to suggest the picking out of a specific individual (i.e. to combine definite and indefinite). All instances are picked up by subsequent personal pronouns.

This should, of course, also be viewed in the context of the pragmatic performative setting within which a state of the union address occurs. An important part of the ritual, in the modern era, has been to invite members of the general public, who are deemed especially worthy and exemplify key points the president wishes to make, to act as rhetorical props. The president refers to them first as ‘a nurse’ or ‘a firefighter’ or ‘a veteran’ and then by their proper name, as the TV camera pans onto them and they stand up to take a bow.

There is a lot going on here ideologically and it crosses party lines. Individuals are seen first, as if from an aerial distance, as an instance of a particular group, until focused on by the media spotlight, whereupon they are addressed by their proper name and blandished with a series of possessive pronouns. Tellingly, however, individuals who are deemed truly important retain their definite article, ‘the president’, ‘the junior senator from Texas’ etc. Describing Ted Cruz as ‘a senator’ and making him stand up in a crowd would be peculiarly demeaning.

More old-fashioned political discourse tended to focus on groups rather than individuals and on an individual’s status, role and duties as a representative of a group. In times gone by, politicians would be more likely to point rhetorically to ‘the farmer’ as a generic category than pick out an individual farmer as representative of the whole.

Whether this change is a sign of progress or decline is more difficult to tell. In a way, it is just a way of perpetuating injustices by dolling them up in more appealing garb. Calling someone ‘a something’ is like giving them a uniform. It might give them a temporary buzz of self-importance and boost their short-term self-esteem; but after long periods of drilling and bullying and being treated as dispensable, they might come to yearn to be regarded not as ‘a’ something but as ‘the’ individual they truly are. The apparently endearing pat-on-the-back of an “a” tends to tilt towards “just a” and there is nothing just about that. The apparently stuffier and more supercilious language of the founding fathers understood that.

For more on this subject, see: