Oodles of Poodles

I don’t know anything about historical linguistics, and the class I’m taking this semester is only slightly rectifying this sad state of affairs. I can only blame myself, though: I’m way behind on the readings. All the time I set aside for this class is spent puzzling over some of the most complex linguistics problem sets I’ve ever encountered. It’s like, “Here are fifty words in three related languages you’ve never heard of; stare at them for ten hours and determine what their common ancestor must have sounded like.” The irony is that if I spent more time doing the reading, I’d probably know enough to be able to spend less time on my homework. But it may be too late to test this assumption.

The weekend is still young, and I am hoping to be productive. Another problem set lies waiting on my desk. But instead of looking at it, I am looking at photos of dogs. Poodle hybrids, to be exact.

Labradoodle Brown.jpgIt all started when my professor remarked in class the other day that his favorite modern example of extension had to do with Goldendoodles. First, though: what is extension? It is a type of analogical change that affects the morphology of a language, or the way that words are put together. In extension, a somewhat rare form used in one paradigm becomes used in another paradigm, one that used a different form previously or had no kind of alternation at all. Here’s an example: you could have the verb paradigm that changes ‘i’ to ‘o’ in the past tense, like ‘drive~drove’. This then becomes ‘extended’ to other, similar-sounding verbs, and then we get ‘dive~dove’ (instead of ‘dived’) and ‘strive~strove’ (instead of ‘strived’)1. Of course, this isn’t a great example because English past tense forms have a complicated history — in fact, I’d say it’s okay to use any of those forms, irregular or not. I’m not a prescriptivist.

The point is, though, that extension begins with one odd formation and gradually spreads to others. So let’s talk about poodles. The first poodle hybrid was the Labradoodle, introduced in the late ’80s by an Australian breeder who wanted to create a hypoallergenic guide dog. It was a cross between a Labrador Retriever and a Standard Poodle (not a new breed per se), and its name was also a cross: specifically, a portmanteau in which the ‘labrad-‘ from Labrador and ‘-oodle’ from poodle were mushed together2.

English speakers are quite adept at coining new words from combinations of old ones. And when something as cute and clever as ‘Labradoodle’ catches on, well, the possibilities are endless. In the early ’90s, the ‘Goldendoodle’ graced the world with its presence: a cross between a Golden Retriever and a poodle. But wait, people (might have) thought: where does the second ‘d’ in ‘Goldendoodle’ come from?

My professor’s argument is that a process of renanalysis occurred after ‘Labradoodle’ came into common parlance. Of course, the original portmanteau used ‘-oodle’ in the blend, but due to the patterns of English phonology, the ‘d’ was reanalyzed as a syllable onset — and thus as the morpheme onset — et voilà: subsequent poodle hybrids used ‘-doodle’ as a suffix instead. Hence, Goldendoodle. Dalmadoodle. Mastidoodle. Saint Berdoodle.

But there’s a catch. Not all poodle hybrids use the supposed ‘-doodle’ extension. For example, the winning poodle portmanteau name in my opinion is the Giant Schnoodle. And my mother has, for quite some time, expressed interest in taking care of a Shih-Poo. (I can’t help but laugh at how earnestly she says it.) We’ve got Bassetoodles, Great Danoodles, Corgipoos, and Cockapoos. Yes, these all exist. And no, they do not all follow the same pattern as our original Labradoodle. Indeed, it looks like there’s more than one way to, er, skin this cat.

So, I went to the Wikipedia of dog breeds and analyzed all of the poodle hybrid names listed here. And then I made some charts (using R, because I need to practice for one of my other classes). Have a look!


As it turns out, of all the hybrid names listed on the website, 50.7% use ‘poo’ as either a prefix (‘Poogle’) or suffix (‘Yorkipoo’), leaving 49.3% to use ‘-oodle’, ‘-doodle’, or ‘-le’ (as in ‘Rottle’, the Rottweiler-Poodle mix). Furthermore, ‘-oodle’ and ‘-doodle’ are used almost equally.

Now I can put on my phonologist hat and think about what conditions would cause someone to favor one affix over another. I marked the phonetic environment in which these affixes occurred. It turns out that ‘poo’ is used as both a prefix and a suffix, and when it is used as a suffix, it most commonly follows a vowel (‘Pomapoo’) or a rhotacized vowel (‘Shar-poo’). Interestingly, ‘-doodle’ also occurs most often after vowels and other sonorant sounds (like /n/, /ɹ/, and /ə/), but ‘-oodle’ is used when the suffix is added after consonants (like /b/, /d/, /f/, /k/, /t/, and /w/).

It didn’t surprise me that alveolar consonants (/t/ and /d/) would condition ‘-oodle’ instead of ‘-doodle’; a doubled alveolar consonant would simply cause one of them to be deleted. But I was intrigued that /n/, a nasal consonant, conditioned /-poo/ only in ‘Affenpoo’ and ‘-doodle’ only in the peninitial3 ‘Goldendoodle’; otherwise, we got ‘Cairnoodle’ and ‘Bolonoodle’ (instead of ‘Cairndoodle’ or ‘Bolondoodle’).

Here’s a (badly-done) chart of ‘oodle’ and ‘doodle’ distributions. If I knew how, I’d reorder and recolor the variables so that consonants and vowels are next to each other, but hopefully this is at least interpretable. Oh, and the question mark represents a schwa (/ə/). Also, some of the colors are wrong. Ugh, this chart is no good. I’ll redo it in the morning.


Anyway…. the theoretical question we are asking is: historically (that is to say, in the past thirty years), was the portmanteau ‘Labradoodle’ responsible for extension and the creation of a new suffix? Perhaps not, since our (very limited) data set show something that is perhaps better described as productive, yet phonologically-constrained blending.

One interesting example still support the case for a ‘-doodle’ morpheme, however. Apparently, the offspring of an Irish Setter and a poodle is called, simply, an Irish Doodle.

Much playful, very linguistics, wow!


Word of the Day: the rumen is the first of the four stomachs of a sheep or cow. These and other grazing mammals, called ruminants, are known for ‘chewing the cud’. This expression underwent semantic shift from a dietary activity to a mental one. Today, if we encounter a difficult problem, we might say that we ‘ruminate’ over it. In this case, puzzles can really be brain food!

– – –

1 Garrett, Andrew. “Paradigmatic uniformity and markedness.” Explaining linguistic universals: Historical convergence and universal grammar (2008): 125-143.
2 The word poodle itself is a borrowing from German ‘Pudelhund’, referring to a dog that was bred to jump through puddles to catch waterfowl.
3 h/t to Eleanor Glewwe for this word, which is the opposite of ‘penultimate’

About Andrew C.

I'm a grad student at UC Berkeley.
This entry was posted in what even is linguistics and tagged , , , , , , . Bookmark the permalink.

2 Responses to Oodles of Poodles

  1. The opposite of penultimate is peninitial! I learned this word from autosegmental tonal analyses. Also, I totally misinterpreted your “post-original”; I thought it meant something like “the word that this blog post was originally about.”


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s