Title text: The worst is the Terrible Twos, when they're always throwing things and shrieking, "forsooth, to bed thou shalt not take me, cur!"
Megan and Cueball are having what could appear to be a typical conversation about her child's ability to learn languages really fast. But the comic mixes up the concept of learning a language and the development of languages over time. The joke comes from a conflation of two different things.
The conventional meaning of language development is the process by which infants begin to talk, that is to understand and produce intelligible speech. The field of language acquisition (sometimes called... language development) seeks to understand how baby humans are able to rapidly comprehend, internalize, and begin producing a new language.
Instead of starting with babbling, the first stage of normal language development, this baby's form of "language development" seems to be the linguistic form: going through all of the theoretical stages of the evolution of the English language, from Proto-Indo-European to Germanic to Old English.
In comparative linguistics and historical linguistics, Proto-Indo-European is a theorized common ancestor of the Indo-European language family. Proto-Germanic is a reconstructed language formerly spoken in Iron Age Scandinavia. It developed out of Proto-Indo-European and is the proposed common ancestor for all Germanic languages. Old English would have developed out of Proto-Germanic. Modern English developed out of Old English with many additions from French (which comes from a different branch of the Indo-European language family).
This parody of language development parallels the discredited theory of recapitulation in embryo development, sometimes expressed as "ontogeny recapitulates phylogeny", in which a developing animal embryo (ontogeny) was once thought to go through stages resembling successive adult stages in the evolution of the animal's remote ancestors (phylogeny). It also plays off of misconceptions about language evolution. Many people assume that ancient languages are more "pure" and "primitive" and that modern languages are more "complex" and "advanced". The comic takes this idea to its logical conclusion by joking that children should successively graduate between historic languages while learning to speak, which is more obviously absurd — it would take years to acquire any of the languages involved.
In linguistics, reconstructed words from proto-languages are commonly marked with an asterisk (*) to show that the word forms are not attested by any historical sources but created as a proposed ancestor word. The baby says the Proto-Indo-European roots that the words "milk" and "please" are derived from. Obviously, the speakers of Proto-Indo-European did not speak in roots, but used words made from the roots, so the way the baby talks does not reflect any stage of development of the proto-language.
Some sounds babies make are hard to interpret. However, humans have a tendency to recognize known things and patterns. They see what they want to see and hear what they want to hear. Thus, a parent familiar with Proto-Indo-European may falsely hear their baby speak Proto-Indo-European by misinterpreting unintelligible sounds.
Perhaps this is an alternate universe where every baby has to gradually develop their language skills along a historical path rather than a child-developmental one, until they reach the ultimately developed modern language of their parents (in this case Modern English).
There have been alleged language deprivation experiments where newborn infants were not exposed to any spoken language in order to find the "natural human language", in the days before ethics review boards would have forbidden such cruel treatments. Such experiments are known today to be a source for psychological problems at least. Alleged outcomes in the apocryphal sources range from the deprived children imitating other sounds in their environment, to them dying.
In the title text, Randall describes a 2-year-old child as speaking in iambic pentameter and in Elizabethan English, a meter and dialect of modern English used by Shakespeare more than 400 years ago. The Terrible Twos are a colloquialism referring to the developmental tendency of two-year-olds to have more temperamental behavior, as the child's developing assertion of autonomy and self-identity clash with other expectations of behaviour, before hopefully acceptably balancing their assertiveness with social normatism. The toddler's quote of "forsooth, to bed thou shalt not take me, cur!" would roughly be equivalent to "Indeed, you shall not take me to bed, you dog!" in less archaic English.
- [Megan and Cueball are looking to the left at a baby with dark hair. The baby sits on the left side of a table in an elevated baby chair.]
- Megan: He's only 1, so he still mostly speaks proto-Indo-European.
- Megan: But we've heard a few Germanic words already, so Old English can't be far off.
- Baby: *Melg- *Pl(e)hk-
- Cueball: They progress so fast!
This was the second comic to come out after the Countdown in header text started.
add a comment! ⋅ add a topic (use sparingly)! ⋅ refresh comments!
Has a small, child-size, stick figure been used before? I did not find a category on explainxkcd. This might be an interesting trivia to add. --220.127.116.11 18:45, 12 January 2022 (UTC)
- There have definitely been kids on xkcd before. For example: 1145: Sky Color (but I'm sure there are others). --NeatNit (talk) 20:04, 12 January 2022 (UTC)
- That is not a baby. There have been several comics talking about or depicting babies, and lots with kids. The kid you showed there has her own category Science Girl and has thus been used alot. --Kynde (talk) 19:57, 13 January 2022 (UTC)
- Other examples are 674: Natural Parenting, 441: Babies and 1650: Baby Kvarts314 (talk)
- Yes nothing odd there, but we could of make a category for comics with babies or mentioning babies, but not like a character page... Could that be relevant? --Kynde (talk) 11:56, 13 January 2022 (UTC)
- I think it would be relevant. --18.104.22.168 12:36, 13 January 2022 (UTC)
- Created: Category:Comics featuring babies --Kynde (talk) 19:57, 13 January 2022 (UTC)
Actually words linguists use when they try to talk in very old languages sometimes sound like the things my little son might say between his first perfectly pronounced single words.--Gunterkoenigsmann (talk) 18:53, 12 January 2022 (UTC)
Someone needs to say “ontogeny recapitulates phylogeny”
22.214.171.124 18:56, 12 January 2022 (UTC)
Looking at Wiktionary, I believe the child is saying "Milk Please" See also Reconstruction:Proto-Indo-European/h₂melǵ- Bpendragon (talk) 18:57, 12 January 2022 (UTC)
Perhaps "milk place"?
Hopefully he won't say the proto-Indo-European word for "bear". 126.96.36.199 19:09, 12 January 2022 (UTC)Pat
- You mean *hrktos? 20:45, 12 January 2022 (UTC)
- Oops. I think a brown one ate my IP address.188.8.131.52 20:49, 12 January 2022 (UTC)
- I think you mean Arth? ;-) --Kynde (talk) 11:56, 13 January 2022 (UTC)
The pace of early stage development isn't necessarily an indicator for continued development pacing. I didn't start Proto-Indo-European until I was almost 2, but had completed full vowel shift before second grade. 184.108.40.206 21:20, 12 January 2022 (UTC)
- I corroborate this. I hadn't made many full sentences in Proto-Indo-European until around 4, but by 3rd grade I had fully changed to modern english. --220.127.116.11 23:12, 12 January 2022 (UTC)
Though the explanation eventually touches on this (perhaps multiple editors got in there and shuffled this nearer the end) I believe it should really have started with something about how Language Development (in a child) is being confused/conflated with Language Development (in human (pre)history). It would get straight to the point, I believe. It could then continue to go the further mile in getting into the deconstruction of it all. I'm leaving it unedited by myself, for now, because it deserves a lot more text-shuffling and refining than I can promise to do myself right now, but putting this idea out there to pique the interest of other possible editors. 18.104.22.168 21:29, 12 January 2022 (UTC)
- I have done this now. Originally I and someone else both submitted a really long description at the same time and my "merge" in my limited time was just to put my text after his. Now that I have more time, I've gone through and tried to weave the two in a more logical way, and have it starting with the basic explanation of the joke. I'm new to contributing at this level so if someone wants to check it over to make sure it looks good, feel free. Levininja (talk) 00:34, 13 January 2022 (UTC)
"Old English developed out of Proto-Germanic. Modern English developed out of Old English with many additions from French..."
According to John McWhorter, English is the product of Germanic tongues (spoken by Angles or Saxons?) creolized with the local Celtic languages such as the ancestors of Welsh and Cornish. That involved a blending of grammar and some vocabulary. Later came pidginizing with Norse speech of the Vikings, where details like case inflections were blurred or lost. Romance borrowings came yet a bit later, with 1066 and all that Norman Conquest business.
McWhorter's Our Magnificent Bastard Tongue: The Untold History of English is perhaps worth a read; hope I haven't mutilated the gist of it too much. 22.214.171.124 01:01, 13 January 2022 (UTC)
I second a mention/explanation of the whole "ontogeny recapitulates phylogeny" idea mentioned above. In biological evolution that turned out to be an error, and it's obviously an error here, too. Mschmidt62 (talk) 02:34, 13 January 2022 (UTC)
If the baby is speaking Proto-Indo-European (with some emerging Germanic) at age 1, and Elizabethan English by age 2, is anyone able to work out by what age they would be speaking our present form of English? --enchantedsleeper (talk) 10:23, 13 January 2022 (UTC)
- Elizabethan English is very close to Modern English on the timeline so by a few months they'll speak our English. They would have already started forming their own vowel shifts and other unknown English innovations by the age of 3, possibly predicting the future of English.126.96.36.199 11:55, 20 January 2022 (UTC)
The awesome subtext of this comic--and I'm not sure if it was intentional--is that English-speaking children really do learn proto-Indo-European words first, then proto-Germanic, then Middle English, then Norman French, and so on. Our simplest words are our oldest words, and Pa/Ma are relatively unchanged from a possible pre-PIE language (because similar sounds appear in Semitic languages). Our most basic words are usually our oldest words. Our pronouns and articles are proto-Germanic, and the next level of complexity are Anglo-Saxon in origin--mostly animal names and common items and actions. Only when we get into complexity do we encounter the French influences, and then newer words or compounds. The reason for this goes back to how languages change. As a new language "takes over" the first words they replace are legal (because who's running the courts), and high-end goods. Then commercial language shifts, and if there's a religious aspect or educational systems set up you'll see that come over as well. The hardest things to penetrate are the furthest from the invading speakers' influence: farm animals (why we use Anglo-Saxon "cow" for the animal but French "beef" for the meat from it, etc.) and finally the home. And yes, when you have a couple of parents interested in linguistics, they really do point out to each other the origin of their children's speech. Uh...should I be admitting this? MGoSeth (talk) 15:00, 18 January 2022 (UTC)
- Well, as far as cow/beef (and sheep/mutton, etc) that was 'distance from influence' in that the (old-style) francophones of the new ruling-classes rarely concerned themselves with the husbandry side and just used their native term for the eventual food-word whilst the long-standing native anglo(saxo)phone farmers got on with the rearing of the livestock with the animal-word. It wasn't a problem to 'penetrate' the usage but likely more a disinclination to even try, leaving a division of language by context.
- In other bits of English, both anglic and francish roots begot 'equal' terms (though at times class-divided as snobbery or inverse-snobbery drove particular speakers to decide to favour particular groups of synonyms) and thus expanded the language with near-duplicates.
- While dialects all over retained words even more niche than the farm/food distinction. Many NE-English terms, both basic and more complex, hold over from the era of Viking incursions and rule before even the francofied Nor(th)mans of Normandy as well as prior Celtic terms that survived the germanic and pre-germanic cultural influxes.
- More obvious in rural situations (at least for presumably archaic pronunciation, e.g. a ewe being a 'yeow') because industrialisation and driven population accumulation in the 'new' cities rather mashed valley-by-valley/dale-by-dale differences together into Mancunian/Bradfordian/etc superdialects (with still a few locality-based specificities, like the Thee-Thar/Dee-Dar divide over a few miles of industrial South Yorkshire or the "We was/I were" verb-(dis)agreement up in parts of the North Riding cities.
- It's been a while since I studied this, so forgive me if I'm out of date with the latest conclusions, but it was always so interesting how they came to conclusions of what artefacts were what particular age of linguistic 'fossil'. 188.8.131.52 17:20, 18 January 2022 (UTC)
- When you think about it, the baby is saying the words for "milk" and "please" in succession, which means it's saying "milk, please" in Proto-Indo-European. The baby is asking for milk.--Yellow Candy 8432 (talk) 18:56, 23 July 2023 (UTC)
The other cool implication is that since children start with PIE first, that it is the oldest language (unless X thousand years ago is within a baby's recollection but X + 1 thousand is not, which would be inconsistent and silly for X > ~0.001). Since PIE is estimated to originate somewhere in the region of 6000 years ago, that means language itself is 6000 years old. (Unintentionally) quite creationistic.
- How do you sign these? Best attempt: 184.108.40.206 04:00, 6 August 2023 (UTC)
- FYI, you sign by using the
~~~~ markup, which adds whatever is appropriate (or configured) for you. I've changed your Best Attempt to be what it would have been, as of that edit, but you don't need to worry about the detail, the wiki sorts that out for you at the time if you invoke it right. There is also a button above the edit area (next to last on the right) that adds
--~~~~ (the "--" does nothing important, markupwise, and isn't necessary) if you can't find the ~ ('tilde') on your keyboard,.for whatever reason, like being hidden behind two 'layout switches' on my on-screen KB (but then a simple / is hidden behind one, where \ is just a long-press character, and this is an Android system, not a DOS one!). Anyway, in case you/anyone else passes by again and finds this at all useful. Not the DOS bit, but before that. ;) 220.127.116.11 08:38, 6 August 2023 (UTC)