McJob And The OED

28 March 2007

Numerous media outlets over the past week have reported that the McDonald’s corporation is up in arms over the inclusion of the word McJob in the Oxford English Dictionary and other British dictionaries.

The OED, which added the word to its online third edition in March 2001, defines McJob as:

An unstimulating, low-paid job with few prospects, esp. one created by the expansion of the service sector.

David Fairhurst, McDonald’s chief people officer in northern Europe, said of the definition, “We believe that it is out of date, out of touch with reality and most importantly it is insulting to those talented, committed, hard-working people who serve the public every day. It’s time the dictionary definition of McJob changed to reflect a job that is stimulating, rewarding and offers genuine opportunities for career progression and skills that last a lifetime.”

McDonald’s says that it is engaging in a public petition to get the OED and other dictionaries to change their definition.

This isn’t the first time that McDonald’s has objected to how a dictionary defines McJob. In 2003, McDonald’s then CEO Jim Cantalupo objected to Merriam Webster’s definition of the term as “a slap in the face” of the company’s employees.

The 11th edition of Merriam-Webster’s Collegiate Dictionary (published in 2003) defines the term as:

A low-paying job that requires little skill and provides little opportunity for advancement.

Now, almost all English-language dictionaries are based on descriptive principles. They don’t define words as they “should” be used. Rather, they define words based on how those words are actually used by professional writers and the public. Lexicographers gather citations of a word’s use, analyze them, and synthesize a definition based on how the word is used in those citations. And if a word is found in a major dictionary, it is almost certain to be in widespread use—or at least have been used by significant writers or speakers. Dictionaries don’t have room to include frivolous words and even on-line dictionaries with no page limits have limited editorial time that is better spent on words that dictionary readers will actually need. So a word like McJob is not included for fun or because the editors have a political point to make.

The term McJob dates to the mid-1980s. Its first known appearance in print is in a headline in the 24 August 1986 Washington Post:

The fast-food factories: McJobs are bad for kids.

There was an earlier use of McJob by McDonald’s as a name for a program to hire the disabled. From the Los Angeles Times of 29 July 1985:

For instance, the McDonald’s fast-food chain recently began a training program for the handicapped in the San Fernando Valley called McJobs. McDonald’s has hired a dozen people after the two 10-week training programs held so far.

The slang sense of the word recorded by the dictionaries is only tangentially related to the 1985 training program for the disabled. They are both based on McDonald’s habit of trademarking terms beginning with Mc-, but there the connection ends. The slang sense is not based on the 1985 use by McDonald’s, but rather on the perceived low status of employment at the fast food chain.

Corporate protests of dictionary entries are nothing new, but usually they are related to whether or not a term is considered a trademark. Trademarks that enter into common use, become “genericized,” lose their status as trademarks and can cost companies dearly. Xerox, for example, is in a continual fight to maintain the trademarked status of its name and does not want the name used as a generic term for photocopying.

Even Wordorigins.org has been subject to corporate lobbying. In our case it was by Google. In this case, Google was quite reasonable and never threatened any type of legal action. They simply objected to the definition of the verb to google as “to search for something on the World Wide Web.” Google wanted the definition to read “to use the Google search engine to search the web.” I happily compromised on “to search for something on the World Wide Web, particularly to search using Google’s search engine,” mainly because it more accurately reflected usage than my original wording.

The McDonald’s web site does claim that McJobs is a trademark, but from the web site there is little evidence that the term is in active use by the company. There are three appearances of the term on the site: in the list of claimed trademarks, in the title of a podcast where it is used in the sense of a job at McDonald’s, and in a reference to the training program for the disabled. But McDonald’s does not seem to be fighting a trademark battle here.

Instead, McDonald’s seems to be engaged in an attempt to bolster their corporate image, but in this case an exercise in shooting the messenger. The problem isn’t the dictionary, the problem is that McDonald’s employs people at very low wages and no fringe benefits to speak of in mind-numbingly dull jobs that impart little or no useful skills for more advanced employment. This business strategy has netted them billions in profits. The problem isn’t that dictionaries are depicting them badly, it’s that McDonald’s is just not a very nice place to work. They could fix the problem, but that would undoubtedly mean a reduction in profits. The situation is one of their own making and if they’re unwilling to change, they shouldn’t complain when others accurately describe it.

Will McDonald’s efforts result in a change in the definitions? It is unlikely. Lexicographers are quite resistant to lobbying efforts. They are continually inundated with entreaties from cranks and intellectual property lawyers about this term and that word. If anything, McDonald’s is only making the dictionary editors look skeptically at any real evidence that the term might be used in a different sense, fearing that any such usages exist because the corporation planted them.

If a publication published a study claiming that McDonald’s food has unhealthful concentrations of fat, sodium, or whatever, McDonald’s would not respond by circulating a petition in protest of the study. No, they would counter the study with evidence that the levels of fat or sodium were not all that high. Or they would add a line of salads to their menu.

They should respond in a similar manner to such dictionary definitions. Instead of a petition, how about collecting citations of McJob being used in a positive sense and submitting the list to the dictionary? Or raising the salaries of its employees and aggressively advertising a program to promote ambitious line workers to management positions? One does not change public perception by changing the dictionary. Rather, one changes the dictionary by changing public perception.

McDonald’s could take a page from how Google approached Wordorigins.org. They disputed my definition on the basis of it not accurately reflecting how the word was actually used. They prompted me to re-examine uses of the verb and I ended up concluding that they were partially right. The verb was used largely, but not exclusively, in the sense of using the Google search engine. As a result, the revised definition could help them in any potential future trademark dispute, albeit perhaps not as much as they might like, and the readers of this site got a more accurate definition of the term. Everybody won.

alright v. all right

24 March 2007

Is alright all right? Or is it an abomination.

Fowler, who despite his being invoked as a prescriptivist icon is usually pretty reasonable in his commandments, rejects alright and seems to be a major source of the objection to the word. In his classic 1926 Modern English Usage, Fowler writes:

all right. The words should always be written separate; there are no such forms as all-right, allright, or alright, though even the last, if seldom allowed by the compositors to appear in print, is often seen (through confusion with already & ALTOGETHER in MS.

But times change and so does what is considered acceptable in standard English. Robert Burchfield, in his 1996 updating of Fowler’s work injects a class distinction into the usage:

all right. The use of all right, or inability to see that there is anything wrong with alright, reveals one’s background, upbringing, education, etc., perhaps as much as any word in the language. Alright, first recorded in 1893 [...] is the demotic form. It is preferred, to judge from the evidence I have assembled, by popular sources [...] It is commonplace in private correspondence, esp. in that of the moderately educated young. Almost all other printed works in Britain and abroad use the more traditional form.

The second edition of the Oxford English Dictionary (1989) simply notes that it is “a frequent spelling of all right.”

Pam Peters in her 2004 Cambridge Guide to English Usage indicates that there are subtleties to alright that are frequently ignored by commentators, and takes a swipe at Burchfield:

The spelling alright is controversial for emotional rather than linguistic or logical reasons. It was condemned by Fowler in a 1924 tract for the Society for Pure English, despite recognition in the Oxford Dictionary (1884-1298) as increasingly current. But the fury rather than the facts of usage seem to have prevailed with most usage commentators since. [...] Dictionaries which simply crossreference alright to all right (as the “proper” form) typically underrepresent its various shades of meaning as a discourse symbol. It may be concessive, as in Alright, I’ll come with you—or diffident, as in How’re things? Oh alright—or impatient as in Alright, alright!. None of these senses are helpfully written as all right, which injects the distracting sense of “all correct.” Those who would do away with alright prefer to ignore its various analogues, such as almost, already, also, although, altogether, always, which have all over the centuries merged into single words. Objections to alright are rarely justified, as Webster’s English Usage (1989) notes, and Burchfield (1996) only makes a shibboleth of it. [...] At the turn of the millennium, alright is there to be used without any second thoughts.

On this side of the pond, things are similarly muddled.

Bryan Garner, in his 1998 Dictionary of Modern American Usage, is categorical in his rejection of the word:

Alright for all right has never been accepted as standard in AmE. Still, the one-word spelling may be coming into acceptance in BrE.

Yet, Webster’s Third New International Dictionary (1961) states:

In reputable use although all right is more common.

The American Heritage Dictionary (2000) sides with Garner, but is somewhat less categorical:

Despite the appearance of the form alright in works of such well-known writers as Langston Hughes and James Joyce, the single word spelling has never been accepted as standard. This is peculiar, since similar fusions such as already and altogether have never raised any objections. The difference may lie in the fact that already and altogether became single words back in the Middle Ages, whereas alright has only been around for a little more than a century and was called out by language critics as a misspelling. Consequently, one who uses alright, especially in formal writing, runs the risk that readers may view it as an error or as the willful breaking of convention.

As usual, however, Webster’s Dictionary of English Usage (1989) gives the most exhaustive treatment in an entry that runs nearly two pages. The conclusion:

Is alright all right? The answer is a qualified yes, with these cautions. First all right is much more common in print than alright. Second, many people, including the authors of just about every writer’s handbook, think alright is all wrong. Third, alright is more likely to be found in print in comic strips, trade journals, and newspapers and magazines than in more literary sources, although it does appear from time to time in literature as well.

Alright is indeed a fairly recent development, appearing only at the end of the 19th century. There are much earlier uses, but these faded from the language long ago. There is the Old English ealriht and Chaucer wrote in his c.1374 Troilus and Criseyde:

Criseyde was this lady name, al right.

After Chaucer, the phrase all right, in whatever spelling, seems to vanish from the language. It pops back up in Shelley’s 1822 Scenes from Goethe’s Faust:

That was all right, my friend.

The merged word first appears in the Durham University Journal from November 1893:

I think I shall pass alright.

Reaction against the merged form kicked up in the opening years of the 20th century and, as we have seen, has not abated since. It is clear, though, that alright is a common spelling on both sides of the Atlantic. The relatively rarer use of alright in print is due almost entirely to proofreaders and compositors, as it is often seen the handwritten manuscripts of printed works, but is absent from the final, published versions. Alright cannot be rightly called an error and there is nothing inherently wrong with the form, but all right can be legitimately preferred for consistency of style or to avoid letters to the editor from outraged readers.

And perhaps the most important factor in determining whether alright is acceptable or not is that the Microsoft spell checker does not flag it as an error. This gives millions the justification to use the term with impunity.

OED Revision of 15 March 2007

19 March 2007

The editors of the Oxford English Dictionary released a new crop of entries for the online OED on 15 March. Most of the new entries cover the range from Prakrit (the vernacular version of Sanskrit) to Prim (a clipping of Primitive Methodist). This range also contains the prefix pre- and a host of new pre- words are included. Perhaps the most notable of these is the airline jargon term pre-boarding, referring the custom of letting the disabled, those with infants, frequent flyers, and first class passengers to board the plane before the rest of the rabble.

In addition to the scheduled updating the Ps, as usual a number of significant new words were added from elsewhere in the alphabet. These include the Pig Latin ixnay, the exclamation ta-da, and from the world of computing the verb to virtualize and the noun wiki.

Also of note is that the editors have discovered that the verb to set no longer has the longest entry in the OED. It has been supplanted by the verb to make, which took the title back in June 2000 when it was updated. To set, however, may stage a comeback and could very well reclaim its position when the editors get to updating the Ss. Rounding out the top ten longest entries are, in order, to run, to take, to go, pre-, non-, over-, to stand, and red.

28 Days

28 February 2007

Why does February have 28 days when all the other months have 30 or 31? The question may not seem to have much to do with word origins, but when you start digging into the answer you uncover a trove of word origins relating the calendar.

The first is the origin of the word calendar itself. In the Roman system of reckoning dates the kalendae, or kalends, was the first day of the month. (The singular form kalend or calend is sometimes found English, but in Latin the word is always plural.) Kalendae literally means accounts, and debts were due on the first day of the month, hence the name.

The Roman calendar had two other named days each month. The nonae, or nones, was the seventh day of the month in March, May, July, and October and the fifth day in other months. The nones, meaning ninth, is so called because it falls nine days before the ides (when you count inclusively). So the ides would fall on the 13th or 15th, depending on the month. The ides was originally the date of the full moon, making the nones the day of the half moon, but at some point these dates were fixed. The etymology of ides is uncertain, but is most likely of Etruscan origin and unknown meaning. The Ides of March is famous because that is the day on which Julius Caesar was assassinated and because of the famous warning, Beware the Ides of March, that appears in the Shakespeare play.

Our modern calendar is based on the Roman one, which originally had ten months of 30 or 31 days each, with a 61 day period between December and March that fell outside the calendar. This gap was presumably because the calendar was chiefly used to regulate planting and harvesting and this period was unimportant to farmers. The later addition of two more months explains why the numerical Latin roots of the months’ names are two off from their position on the calendar. October was originally the eighth month, September the ninth, etc. The original Roman months were:

• Martius (31 days), named for the god Mars

• Aprilis (30 days), after Apru, the Etruscan name for Venus

• Maius (31 days), after the goddess Maia

• Junius (30 days), after the goddess Juno

• Quintilis (31 days), the fifth month

• Sextilis (30 days), the sixth month

• September (30 days), the seventh month

• October (30 days), the eighth month

• November (30 days), the ninth month

• December (30 days), the tenth month

The months of January (Ianuarius), named for the god Janus, and February (Februarius), from a Sabine word meaning purification because of religious rituals conducted during that month, were added during the reign of Numa Pompilius, the second king of Rome (c.715-673 B.C.), bringing the entire year within the calendar. But in Roman reckoning, these were the last two months of the year, not the first. The Roman new year began in March.

Under Pompilius’s calendar, the months that previously had 30 days were reduced to 29. Ianuarius also had 29 days and Februarius had 28. In alternate years an extra month, Mercedonius, was added at the end of February to align the calendar with the solar year. In such years February was cut short to 23 or 24 days, to create an average year of approximately 365 days.

Unsurprisingly, this complex calendar was difficult to maintain. Finally in 45 B.C. Julius Caesar implemented a major calendar reform. He, or more accurately his astronomers, added ten days to the calendar year, bringing it into line with the solar year and eliminating the need for the biennial intercalary month of Mercedonius. He set the length of the months to those we’re familiar with today and instituted the practice of the modern leap day, an extra 29th day of February every fourth year. During the reign of Augustus, the month of Quintilius, the month of Julius Caesar’s birth, was renamed Iulius or as we call it today, July, and Sextilis was named Augustus after the current Emperor.

There is a legend that July was given 31 days, with the extra day taken away from February, so that no month would be shorter than the one named after the great Julius. And not to be outdone by his adoptive father, Augustus made his month 31 days long as well, again taking the day away from February. It sounds good, but it’s not true. July and August both had 31 days and February 28 before the former two were renamed after the Roman leaders.

This Julian calendar was a significant improvement, but it was not perfect. The most significant problem is that the average Julian year of 365.25 days is 10 minutes, 48 seconds too long. The earth actually takes 365 days, 5 hours, 49 minutes, and 12 seconds (or 365.2425 days) to orbit the sun. This may not seem like much, but over the centuries the error accumulated.

As a result, the vernal equinox slowly drifted to an earlier date each year. This caused a problem for the Christian church, which calculated the date of Easter based on the equinox. On the command of Pope Gregory XIII, the calendar was revised in 1582. Days were skipped, returning the equinox to its traditional date. And to keep the calendar on track in the future, century years skipped the leap day on 29 February unless the year is divisible by 400 (1600 and 2000 had leap days, but 1700, 1800, and 1900 did not). This new Gregorian calendar also moved the new year to the beginning of January.

The Gregorian calendar was not implemented all at once. Catholic countries tended to adopt it early, Protestant ones later. Great Britain and its American colonies did not adopt it until 1752. Russia did not adopt the Gregorian calendar until after its revolution of 1918. If you do genealogical or historical research, you will often see dates prior to 1752 marked as either O.S., old style, or N.S., new style. O.S. dates are those of the original Julian calendar. Those marked N.S. have been adjusted to match our modern Gregorian calendar. George Washington, for example, was born on 11 February 1731 O.S. according to the Julian calendar which was in force in Virginia at the time, but today that date corresponds to 22 February 1732 N.S. The shift of 11 days is due to the skipping of days in order to return the equinox to its traditional date and the shift in the year is due to the move of New Year’s to January.

The Gregorian reforms brought the calendar very close to the actual solar year, but it is not quite accurate. Modern astronomers periodically add leap seconds to the clock and calendar to keep them in line with the earth’s rotation around its axis (which is not precisely 24 hours) and its orbit around the sun. Most of us don’t even notice when this is done.

See what you get by asking a silly little question like why does February have 28 days?

Holidays of the Season

22 December 2006

This Monday is Christmas, the biggest, albeit not the most theologically important, Christian holiday. As I’m sure you know, Christmas is the celebration of Jesus Christ’s birth. Tradition gives this date as 25 December, although no one knows what day he was actually born. The December date is believed to have been selected because it corresponds with a variety of pagan festivals celebrating the winter solstice, most notably the Roman festival of Saturnalia, honoring the god Saturn. Co-opting the traditions and festivals of other faiths is a time-honored religious practice.

The word Christmas literally means Christ’s mass. The modern form comes from the Old English Cristes mæsse, which is first recorded sometime before 1123 in the Anglo-Saxon Chronicle for the year 1101:

Her on thisum geare to Xpes. mæssan heold se cyng Heanrig his hired on Westmynstre. (In this year at Christmas held the King Henry his court in Westminster.)

Many object to the abbreviation Xmas, meaning Christmas, because it “takes the Christ out of Christmas.” But this is actually not the case. The X is not a generic substitution of a letter for the name of Christ, but rather it is the Greek letter Chi, the first letter in the Greek spelling of Christ. As we can see in the above quotation from the Anglo-Saxon Chronicle, this use of Chi as an abbreviation for Christ dates back to the Old English period and is not part of the modern commercialization of the holiday. Here is another example from one of John Wyclif’s sermons c.1380:

In this word Vix ben but three lettris, V, and I, and X. And V bitokeneth fyve; I bitokeneth Jesus; and X bitokeneth Crist.

The abbreviation Xmas itself dates to the mid-16th century. A 1551 example found in Edmund Lodge’s 1791 Illustrations of British History:

From X’temmas next following.

And we have this one in the modern form from c.1755 found in Bernard Ward’s 1893 History of St. Edmund’s College, Old Hall:

In ye Xmas and Whitsuntide Vacations.

We often see the word yule used to represent the winter holidays. Yule comes from an Old English name, geól, the name of the months we call December and January. Bede, writing in De Temporum Ratione in 726 attests to the word in this sense. Bede wrote in Latin, but this text is the only extant documentation of the Anglo-Saxon calendar and does attest to the word’s existence:

De Mensibus Anglorum�Primusque eorum mensis, quem Latini Januarium vocant, dicitur Giuli. (About the English months�The first month, which in Latin is called Januarium, is Yule.)

The use of yule to refer to Christmas dates to sometime before the year 900 when the word is used in this sense in the Old English Martyrology:

Feowertig daga ær Criste acennisse, Thæt is ær geolum. (Forty days before Christ’s birth, that is before yule.)

In Britain, the day after Christmas is called Boxing Day. Traditionally on this day the upper classes would give boxes of gifts to servants and trades people. The name either comes from this practice or from the practice of churches to open the alms boxes on this day and distribute money to the poor. The term Boxing Day dates to at least 1833 when it appears in The Memoirs of Charles Mathews, Comedian:

To the completion of his dismay, he arrives in London on boxing-day.

There are other explanations for the name giving various medieval practices as the source. But given that the term only dates to the 19th century, these would seem to be unlikely origins.

Of course Christianity is not the only religion to have a holiday during this season. The Jews celebrate Chanukah, also spelled Chanukkah and Hanukkah. The name is the Hebrew word for consecration and the festival celebrates the rededication of the Temple in Jerusalem by the Maccabees after the revolt against the Seleucids in 165 BCE. The celebration lasts for eight days beginning on the 25th of Kislev on the Jewish calendar, which falls in November or December (or rarely in January) on the European calendar.

Many African-Americans celebrate Kwanzaa, a secular holiday that emphasizes the people’s roots in
Africa. The name comes from the Swahili kwanza, meaning first, as in the Swahili phrase matunda ya kwanza, meaning first fruits of the harvest. The extra A was added to round out the word to seven letters, which represent seven principles, unity, self determination, collective work and responsibility, cooperative economics, purpose, creativity, and faith. The celebration was started in 1966 by Ron Karenga, an African-American writer and political activist.