Linguistic change on a short time scale

« previous post | next post »

From Reuben Fischer-Baum, "Six Decades of the Most Popular Names for Girls, State-by-State", Jezebel 10/19/2013:

Someone should take a fuller slice of the historical data and calculate the latent regional affinities that this dataset implies. Or maybe someone has already done that?


  1. Yerushalmi said,

    October 20, 2013 @ 7:14 am

    The weirdest thing about this graph is how Jennifer, which looks like it was in first place in just a single state in 1969, jumped in just one year to being the top name in the country, and managed to take over the entirety of the country from 1973-1978!

  2. GeorgeW said,

    October 20, 2013 @ 7:22 am

    Is there a way to slow or stop the yearly progression?

    In a quick glance at a moving target, it looks like a number of the names spread from what I would think to be unlikely states in the Mid-west and South. I would have expected a more coastal and urban genesis.

  3. Tom S. Fox said,

    October 20, 2013 @ 7:49 am

    In what way is that linguistic change?

  4. Ellen K. said,

    October 20, 2013 @ 7:53 am

    GeorgeW, if you go to the linked page, you can see each year individually.

    [(myl) Or you can go to the linked SS website and get the ranked lists by year.]

  5. Tom Scocca said,

    October 20, 2013 @ 8:52 am

    Yerushalmi: I was born in 1971 and grew up surrounded by Jennifers, and my understanding was that it was because of Jennifer Cavilleri in Erich Segal's Love Story. The novel hit the bestseller list in February 1970, and the movie was in movie theaters by December.

    [(myl) The nationwide numbers show a more gradual shift than this story seems to predict:

    I'd interpret this plot as showing that Segal's choice of name was as much the result of a trend as the cause of a trend.]

  6. Mara K said,

    October 20, 2013 @ 9:28 am

    @Tom Scocca: What about "Isabella"'s moment in the sun? That seems to correspond neatly with the end of the Twilight saga.

    [(myl) Here's the overall count for Isabella from 2000-2012:

    The rise from 2000 through 2005, when the first Twilight novel was published, again suggests that Stephenie Meyer's choice of name was as much the result of a trend as the cause of a trend. The decline after 2010 might be related to the publication of the last novel in that series, but Twilight movies continued to be released in 2011 and 2012…]

  7. bloix said,

    October 20, 2013 @ 9:31 am

    Anyone interested in this topic would like the Baby Name Wizard, The Name Voyager is a great interactive graph and is good for hours of fun.

    Just a few tidbits: One big story in girl's names over the past century or more is the long-term dominance and then steady decline of Mary: for decades from the 1880's well into the 20th c, over 25,000 girls per million were named Mary, compared to 1,000 today.

    Another big story is the rapid rise and then fall of Jennifer (17,000/million at its peak) and Jessica (12,000/million at its peak). Emma, Sophia, Madison and Ashley never came close to these numbers.

    And maybe the biggest story is the proliferation of names – there are more different girl's names in common use now than at any prior time.

  8. Tom Scocca said,

    October 20, 2013 @ 9:35 am

    Yup, that would kill off Love Story as anything more than an accelerant. This chart seems to imply some sort of zero-sum relationship between "Lisa" and "Jennifer"–that the girl who would have been named Lisa in 1965 was the girl who got named Jennifer in 1975.

    [(myl) Well, the number of girl babies per year is roughly constant, so the total distribution of names to infants is a zero-sum process; and to the extent that there are a few big winners each year, an increase in one of them will tend to correspond to a decrease in another. It would be interesting to see what sorts of random processes would model the patterns of changes well — perhaps someone has done this.]

  9. bloix said,

    October 20, 2013 @ 10:05 am

    Isabella came from nowhere to begin its rise in the early 1980's and steadily increased in popularity to 2008, when it took a steep jump upward (from 4400/million to about 5800/million) in 2010, and then fell back. So the last bit may be related to Twilight but overall the rising popularity predates Twilight be a couple of decades.

    Isabelle shows a similar pattern but with much lower popularity (topping out at 950/million).

  10. bks said,

    October 20, 2013 @ 10:13 am

    Donovan song Jennifer Juniper was released in 1968.

  11. bloix said,

    October 20, 2013 @ 10:23 am

    "to the extent that there are a few big winners each year, an increase in one of them will tend to correspond to a decrease in another."

    This isn't necessarily so. Although the number of girl babies is roughly constant, the number of names is not. The rise of a popular name could cause a decline in another popular name, or it could cause a decline in the incidence of a number of much less popular names. Decades ago there were fewer names each with larger shares; now there are more names, and only a few reach share levels that were common in prior generations.

    Broadly speaking, there are two different trends at work in girls' names today: one is the rapid rise and fall of very popular names, and the other is the introduction of many more unusual names – whether newly invented, resuscitated from decades past, adopted from traditional boys' names or surnames, or introduced from other cultures. It's hard to know how these trends interact with respect to the popularity of any individual name.

    I would be hesitant to draw any conclusion about the 1975 naming practices of the sort of family (cultural, socio-economic, geographic) that named its child "Lisa" in 1965 without a lot more data than the two-name graph showing a cross-over with Jennifer.

    [(myl) Right — that's why I think it would be interesting to check the fit of various random-process models. Depending on its assumptions and the facts of the case, a model might well fail to provide a good picture of the observed dynamics — and the fact that failure is possible means that it may be scientifically interesting to know which models fit and which don't.]

  12. Morgan J said,

    October 20, 2013 @ 10:41 am

    The 'Lisa' revolution begins in Nevada and 'Jennifer' emerges from Utah to conquer the nation. Anyone know what people in the Great Basin are naming their kids today?

    [(myl) The top 10 girl's names in Utah in 2012 were Emma, Olivia, Sophia, Ava, Lily, Elizabeth, Abigail, Brooklyn, Lucy, and Claire. Nevada led with Sophia, Isabella, Emma, Olivia, Mia, Emily, Ava, Madison, Chloe, and Abigail. The nationwide top ten were Sophia, Emma, Isabella, Olivia, Ava, Emily, Abigail, Mia, Madison, and Elizabeth.

    So maybe "Brooklyn"? Doesn't seem likely, but …]

  13. Jonathan Mayhew said,

    October 20, 2013 @ 11:05 am

    I remember the Ashleys of Indiana.

  14. JW Mason said,

    October 20, 2013 @ 1:28 pm

    I'mm wondering if there is any consistent pattern of regional priority — are there consistent leads and lags in the popularity of names across states? In some cases it sort of looks like names diffuse outward from Appalachia, which is interestingly different from what we normally think is the main gradient of cultural change. But that's just on casual examination — to answer the question systematically, you'd want to look at the full list of names and not just the top ones.

  15. Marek said,

    October 20, 2013 @ 1:48 pm

    Coming soon: Quentin D. Atkinson uses a statistical model adapted from virology to support the Great Basin Origin hypothesis, then points to deep girl name ancestry across the world by looking at ultra-popular names.

  16. ItalianFoodie said,

    October 20, 2013 @ 1:49 pm

    I'm a Carol born in 1952. Whenever I run into another Carol, she's my age and it's kind of fun to talk to someone with roughly your own experience of years. Maybe the Jennifers and Ashleys will feel the sane way.

  17. Eric P Smith said,

    October 20, 2013 @ 2:44 pm

    What strikes me is how unfamiliar I am with so many of these names. I'm British and I don't think I personally know anyone called Madison, Samantha, Alexis, Ava, Olivia, Melissa or Brittany. I'm fairly sure I've never even heard the name Ava. I don't know if that's a US/UK thing or an age thing or a social deficit on my part: probably all three.

  18. bloix said,

    October 20, 2013 @ 3:24 pm

    Eric, I don't believe you never heard of Ava Gardner. How old are you?

    I work with English clients and watch a fair amount of UK television and I'm often struck by how different the English names are. Philippa is my absolute favorite – it's exotic and proper and upper-crust and sexy and utterly absurd all at the same time.

  19. Marja Erwin said,

    October 20, 2013 @ 3:52 pm

    Where is Sara/Sarah in this?

  20. Tracy W said,

    October 20, 2013 @ 4:12 pm

    Bloix: how funny. I think of Philippa as a fairly boring girl's name, yet another one that is just an 'a' stuck onto the end of a boy's name. There were a reasonable number of Philippa's around when I went through school in NZ (none of them boring I admit, but none of them upper-crust either).

  21. Sili said,

    October 20, 2013 @ 4:27 pm

    The colourscheme really annoys me. Why not stick with one colour per name – at least when that name repeats from year to year. Tufte must be spinning in his grave.

    I went at checked "Brandon" when I first saw this the other day, and as expected it grew significantly in the 90es. Nothing similar for "Brenda", though.

  22. Eric P Smith said,

    October 20, 2013 @ 5:45 pm

    @bloix: I'm 64. No, I've never heard of Ava Gardner. You may laugh. I've looked her up now. I have a very uneven cognitive profile typical of Asperger's Syndrome, and film means nothing to me because I have specific difficulties with both story and faces. But I have an excellent auditory memory and I'm sure I would have remembered the name Ava Gardner if I'd ever heard it.

  23. Alexander said,

    October 20, 2013 @ 7:16 pm

    @Sili : There was a massive "Brenda" spike circa 1948-62, such that in the 70s you couldn't rollerskate once around the rink without hitting 2 or 3 Brendas.

  24. Jerry Friedman said,

    October 20, 2013 @ 7:20 pm

    I've known three sets of parents who thought they were giving their children unusual names and hit names that were among the most common then. I wonder how frequent that is. (Much less now that you can just look up last year's most common name, I suppose.)

  25. Jason said,

    October 20, 2013 @ 7:28 pm

    The Madison outbreak was a lot more contained than I thought. Gene Weingarten wrote a truly epic kids today (or rather, parents today) rant about that one.

    (Myself I quite like Madison, even if every single Madison in existence today can be directly derived from the name of the mermaid in that Darryl Hannah movie.)

    Rather disappointed about the ascendance of "Sophia", which would have been my first daughter, but if it's the cliché of the moment, I guess I'll just have to go with little Jerrica, Jem and Synergy.

  26. Joshua said,

    October 20, 2013 @ 7:47 pm

    @Marja: Sarah has had relatively consistent popularity, at least by the standards of girls' names, although it has gone up and down. For the last 50 years, it's remained in the top 100 girls' names every year. It was a top 10 name every year from 1978 to 2002, peaking at #3 in 1993. Last year, it was at #43.

    Sara peaked at #26 in 1991. Last year, it was at #143.

  27. J.W. Brewer said,

    October 20, 2013 @ 8:07 pm

    The decreasing concentration (what percentage of total newborns collectively get one of the N most common names) of popular names needs to be a major part of any analysis of these changes over time. The 2012 "winner" (i.e. most common) girls' name was Sophia, but its absolute popularity (percentage of girls given the name) would have only ranked it 13th (behind Barbara but just ahead of Brenda) in 1960.

  28. Shirley Steele said,

    October 20, 2013 @ 8:13 pm

    Does anybody know the time and geographic scope of the "Linda" phenomenon? I was born in '47 and all those girls that you would run into at the skating rink in the late '50's were named Linda.

    [(myl) Time is easy — and 1947 was indeed Peak Linda:

    Geography would take more work than I have time for this evening.]

  29. Eric Banks said,

    October 20, 2013 @ 8:24 pm

    There's an amusing running joke in Robert Altman's "California Split" (1974) that many of the female characters are named Linda. If memory serves, it's one of the final lines ("They're all named Linda") of the movie as well.

  30. Roger Lustig said,

    October 20, 2013 @ 9:19 pm

    Linda…huge spike in 1947…that's when the song "Linda" (actually written for the infant Linda Eastman) became a hit.

  31. Pat said,

    October 20, 2013 @ 9:46 pm

    This post reminds me of a chapter in the book Freakonomics. Maybe some of you have read it, but for those of you who haven't: the authors of that book said basically what Mr. Liberman has already said a few times here, which is that often the name was already becoming popular before the celebrity, celebrity baby or fictional character had the name. Obviously that could be why they were given the name.

    I thought that chapter was the most interesting one in the book. The way they described the cycle baby names go through reminded me of the cycle of some linguistic changes I have read about, particularly R-dropping (and later "R-restoration") in New York City. Namely, there seems to be kind of a "lower class lag", if you will, with both of those things. I don't understand why that is though.

  32. Bloix said,

    October 20, 2013 @ 11:46 pm

    Eric P Smith – no, I'm not laughing – it's just that Ava is one of those names that is strongly associated with a single famous celebrity. But you may be interested to know that Ava is zooming up the charts as a favorite name in the UK, so if you ever find yourself in a roomful of babies there might be an Ava in there with you.

  33. R. Sabey said,

    October 21, 2013 @ 4:20 am

    @Eric Some of those names were popular here in Britain, too, just not necessarily at the same time they were popular in the USA, that's all. Samantha was 9th most popular girl's name in England & Wales in 1974 and 8th in 1984. Olivia, already 24th in 1994, was 1st in 2008-2010.

    One name that was most popular in the UK earlier than in the USA is Elizabeth, a 1940s-60s name here but in the top 10 in the USA only from 1980 to 2008.

    @Marja Erwin Sarah was in the top 6 from 1979 to 2001, but made 3rd only in 1993, and never got higher. Just one of those names that never quite made 1st, e.g. Karen, Kimberly, Michelle, Amy, Melissa, Amanda.

  34. dw said,

    October 21, 2013 @ 12:04 pm

    The leader in adopting each name was:*

    Lisa: NV
    Jennifer: UT
    Jessica, ME, VT
    Ashley: IN, SC, TN
    Emily: CT, ME, NH, VT
    Emma: IA, MN, NE, VT
    Isabella: CT, CO, FL, HI, NJ, NM, RI
    Sophia: AK, DE, MD, MN, ND, NH, OR, SD, VT, WA, WI

    I hereby declare Vermont the national leader in nomenclature.

    *apologies for any errors

  35. Margaret Dean said,

    October 21, 2013 @ 12:17 pm

    If "Sarah" counts as a different name than "Sara," that may explain why "Brittany" never hit the top spot. I remember lots of girls with that name in my son's generation (they were born in the early 90s), but there were as many different spellings as there were girls, it seemed like! Same goes for all the variations of "Catherine."

    Is there a similar site for boys' names, just out of curiosity?

  36. dw said,

    October 21, 2013 @ 1:56 pm

    Here's my dime-store hypothesis:
    * In the pre-Isabella era (which corresponds neatly with the pre-Obama era in politics), naming fashions were driven by non-Hispanic whites. Other groups would copy the latest fashions later, if at all.

    * Hence the states with high proportions of non-Hispanic whites tended to be in the vanguard of naming trends.

    Here's the list of state leaders in adopting each name 1above, followed by their rank in whiteness as of 2008.

    Lisa: NV #30
    Jennifer: UT #10
    Jessica, ME #2, VT #1
    Ashley: IN #18, SC #43, TN #27
    Emily: CT#26, ME #2, NH #3, VT #1
    Emma: IA #6, MN #14, NE #13, VT #1

  37. Cleveland Evans said,

    October 21, 2013 @ 3:25 pm

    As the above messages about Sarah/Sara and Brittany point out, one of the things you have to remember about this data is that it counts every different spelling as a separate name. I am sure that there would have been some years back in the 1970s when Sarah/Sara would have been the #1 name in some states if both spellings were counted together. Ashley would also have been the national #1 in more years if the alternative spellings like Ashlee, Ashleigh, Ashlea, etc. were added in.

    The influence of pop culture on names often shows a "feedback loop" effect. A name will begin to rise and already be fairly common when a television or film character will get the name and it will be further "goosed". Screenwriters often seem to choose names for adult characters from among those presently popular for children rather than from among those really common for adults the age of the character.

    Emma is an example of this. It was already steadily rising at a good rate when the baby on "Friends" was named Emma and the name took another sudden jump. Emma may well have eventually reached #1without "Friends", but I think it would have been several years after if it hadn't been for the extra push from the TV influence.

    Isabella is another good example because Stephenie Meyer has stated she named the character in "Twilight" Isabella because it's the name she would have given a real daughter if she'd had one. (She has only sons.) So she gave her teenage character a name that was common for babies and the name then got another little boost from the character. Just the Hollywood feedback loop at work again.

  38. J. W. Brewer said,

    October 21, 2013 @ 4:59 pm

    I am particularly impressed by the stretch 1996 through 1999, when each of the "Four Corners" states (UT/CO/AZ/NM) had a different #1 name from the other three. Note that if you swap in for the table dw linked to, the related-but-different table of rank in "non-Hispanic whiteness," New Mexico is the second-least-non-Hispanic-white state in the country and does indeed intermittently stick out from its neighbors or even the nation at large in the temporal sequence of most-popular-girls'-name maps.* Hawaii is the champion in that dimension (i.e. has the nation's lowest %age of non-Hispanic whites) and also seems an outlier in some years. OTOH, in a year like 1996 where the Hawaiian preference for "Taylor" is shared only by Wyoming and South Carolina, it seems rather hard to come up with a demographic or cultural explanation that would unite those 3 states against the remaining 47.

    Interestingly enough, dw lists HI and NM among the joint innovators of Isabella-as-#1, but as part of a larger set with 5 states that lack any obvious common theme.

  39. J. W. Brewer said,

    October 21, 2013 @ 5:17 pm has some interesting data: how do the most popular 100 baby names for California compare to the national list? The drivers of some of the differences are obvious (e.g. California has a significantly higher percentage of Hispanic newborns than the nation at large, so boys' names like Jose/Juan/Jesus that remain uncommon among the non-Hispanic population unsurprisingly rank higher in California); others perhaps less so. I suspect that it's quite unusual for a state to have a #1 name that's not reasonably popular (top ten, let's say) nationwide, but as you go down the charts a bit you might be more likely to hit names where the degree of state-to-state variation (possibly driven by different ethnic/racial mix, but there could well be other factors at play) is more dramatic. FWIW, it should be kept in mind that the relevant data on ethnic/racial mix would in any event not be that for the state population as a whole, but that of the state's newborns, which is not necessarily the same thing.

  40. Jennifer Jarratt said,

    October 22, 2013 @ 8:26 am

    I was named Jennifer in the 1940s (UK) when it was rarely used and considered quaint (esp. in Italy for some reason). Think the connection then was the movie star Jennifer Jones. Now of course…..

    For those of us out there, the name in ancient languages means "White wave."

  41. Bloix said,

    October 22, 2013 @ 8:02 pm

    Tracy W – I've never actually met a Philippa. But when I was driving around Europe with my family we set the GPS on "female – British accent" and called it Philippa the whole trip ("as in, "ask Philippa how to get there.")

  42. Barbara Phillips Long said,

    October 22, 2013 @ 11:02 pm

    I would like to see some charts about the use of punctuation in names, particularly African-American names. I don't know if there's a relationship between the increased use of apostrophes and AAVE. There's a discussion of apostrophes and dashes in names here:

    The post says that the Social Security database doesn't include internal punctuation or capitalization, which is unfortunate because it means the data is flawed.

    There's also been some interesting speculation about hurricane names and their phonological influence on naming trends:

    For a list of retired hurricane names, go here:

    The emphasis these days on names that begin with vowels, particularly A, could be influenced by a number of storm events, including Hurricane Agnes in 1972, Hurricane Allen in 1980, Hurricane Alicia in 1983, Hurricane Andrew in 1992, and Tropical Storm Allison in 2001. I named a daughter Allison in 1979, but spelled it differently based on a spelling I had seen elsewhere, only to discover that there were little Allisons her age all over the place.

    It might be interesting to find out if parents are now discarding some name choices because Internet research shows the name is popular. Maybe we have already passed some "peak name" event, given that ubiquitous names such as Mary have fallen deeply out of favor. I know I had planned since high school to name a son Brian, and gave up on the name because it had become so popular. Imitation continues to inspire parents, however, because there seem to be a lot of Nevaehs around (Nevaeh is heaven spelled backward).

  43. Barbara Phillips Long said,

    October 22, 2013 @ 11:08 pm

    @Margaret Dean:

    Bloix linked to Baby Name Wizard. You can track boys names and girls names there. They have popularity graphs, U.S. maps of the distribution of particular names, and more:

  44. Joshua said,

    October 23, 2013 @ 1:20 am

    @Margaret Dean: The Social Security Administration compiles the statistics on boys' names as well as girls' names. See and look around the site.

    Naming trends for boys don't seem to change as much as for girls. For example, Michael has been one of the top 10 boys' names every year since 1943. That includes a 38-year streak from 1961 to 1998 in which it was the #1 name every year. By contrast, the longest current top-10 streak for a girls' name belongs to Emily, which has been a top 10 name only since 1991.

  45. David P said,

    October 23, 2013 @ 7:36 am

    Ava, or at least Hava, goes way back, although Social Security records don't go back that far. For a while, according to one authority, it was at 100% among girl's names.

  46. dw said,

    October 23, 2013 @ 9:06 am

    @David P

    … but was she named as a girl or a fully grown woman? :)

  47. J. W. Brewer said,

    October 23, 2013 @ 11:16 am

    The spelling "Eva" was in the top 100 in the U.S. from the 1880's when the records begin through the early 1930's. It then began a slow decline, bottoming out at #367 in 1997 before enjoying a renewed surge of popularity (presumably on the coattails of the "Ava" boom), and reentering the top 100 in the last few years. The spelling "Eve," which you'd think ought to be the normative/default one in an Anglophone culture, has been for whatever reasons much rarer in the U.S. as an actual given name over the last 130+years. One verse in the Septuagint uses "Zoe" as a Greek calque for the Hebrew name Hava (obviously subject to a plenitude of alternative transliterations . . .), but it's not clear to me to what extent that connection has had much impact on the recent vogue for Zoe as a girls' name among Americans not of Greek ethnicity.

  48. Mark said,

    October 23, 2013 @ 1:31 pm

    It looks to me like you could pick out chains from the more modern data. Like Emma fading as Eva rises (as though hundreds of moms are saying "There are too many Emmas out there but I really like the name so…") and tons come up with Eva over the next few years. I wonder if there aren't 2 or 3 "chains" of trends overlapping where new popular names are echos of or responses to prior names being popular.

    Angela to Amanda to Samantha, Emily to Emma to Eva, Ashley to Brittany an such.

  49. Changing Norms » No Contest Communications said,

    October 23, 2013 @ 3:49 pm

    […] The fine folk at Language Log dig into the data with their typical, amusing obsessiveness. Share this:Share on TumblrEmailPrint  Posted by nocontest at 1:49 pm  Tagged with: language […]

  50. Ken Brown said,

    October 23, 2013 @ 5:15 pm

    @Jennifer Jarret – I was born in England in the late 50s and Jenny was one of the most common names in my age cohort at school and university. As were Sue and Liz.

    I assume they had Jennifer, Susan, and Elizabeth on their birth certificates, but of course I don't know.

    A few years later we had a wave of Sarahs, Sam(antha?)s, and Tracys.

  51. Ron said,

    October 24, 2013 @ 12:23 pm

    I noticed that Donna showed up in NJ in 1960. Hurricane Donna did, too. Coincidence, or can we expect a bunch of Sandys?

  52. Anthony said,

    October 29, 2013 @ 11:42 am

    My own name has become more and more popular, and last I looked, Anthony is #1 in Nevada and #2 in California, while still around #10 nationally. There's probably a thesis or three in why Hispanic parents are naming their children Anthony instead of Antonio (the trend seems to be driven by Hispanics, not blacks or whites), even though they're naming their kids Jose and Juan instead of Joseph and John.

RSS feed for comments on this post