The Awful German Language (1880)

This is true, but some German compound words acquire a meaning that doesn't simply derive from their component words. Well-known ones include Kindergarten and Weltschmerz. This is often the case for domain-specific terms (Gestaltpsychologie, Bildungsroman).

It's true that English uses basically the same method to create compound nouns, but quantitatively it's a difference. Long compounds consisting of 3, 4 or more parts are completely common in German and cause usually no trouble in understanding, whereas English is far more likely to split them up by the introduction of words such as "of", "for", etc.

x100 this. You can sort of derive the meaning of a complex word if you grasp one or two parts of it and offer a hacked together English translation, even if it doesn’t map directly. I find that people online who haven’t actually studied German like to meme this often.

The Latin-derived cases from the article, on the other hand, are the truly maddening, and makes you appreciate the simplicity of English grammar by comparison.

This is so true. My favourite example is when Top Gear made fun of the German word "Doppelkupplungsgetriebe" by spelling it, when it is quite literally the translation to "dual-clutch transmission". It stil is hilariously funny, but you cannot conclude that German is weird with these words.

In some way yes, but not really. Had a colleague working on a project for Deutsche Bahn (state owned train operator), he was developing an app and the domain knowledge was full of long German words, no one outside this bubble ever uses: Bremshundertstel, Bremszettel, Mindestbremshundertstel, Notbremsüberbrückung... In a way it's better to have a super long name for this, so there are no 2-3 ways to describe the same thing.

I wonder 'where' these compound words end up in an n-dim embedding space (relative to their German and say English 'parts'). In fact this brings up the interesting question of tokenization of the long German compound words, and how all this plays out in German to English (and reverse) LLM translation and text generation.

Sure, you can say three nouns in a row in English. But can you then make them into a verb? Or and adjective? What happens when some of the three words in English already are in a form that also parses as a verb or an adjective?

English is a bastard language and it shows in its grammar.

My favorite example is "kartoffellerntepause", it's the German word for the school break in southern Idaho for potato harvest.

Yeah nothing worse than entering a translated to English portal for Dutch tax purposes. Because those English words also ended up in Business Dutch but then got another meaning. Dutchlish, or at least the original term in parenthesis) is really preferable to anything else.

Same as a friend of mine who works for NAV. There's a whole lot of long-ass variable and function names because they use the Norwegian name for whatever they are calculating. It makes sense for them though, as the ones who review your code are lawyers...

I've had this same discussion with a colleague at my job in the Netherlands. He will ask me to choose from a list he provides for variable names. Usually I need to ask for more context and then I end up leaning towards the more "well known/normal" option, both because it still fits and will be more likely to be understood another decade from now when we've probably both moved on and are not there to answer anymore questions.

Discussing the words is a fun way to take a little break during the workday, but I don't consider it more than that.

There very much was a well defined distinction between safety and security: Sicherheit and Schutz, as in Datensicherheit and Datenschutz.

And yeah, you can see with those two latter terms where the issue lies :)

Those two were traditionally actually used this way in the safety and security context - I think I even have the script for the "Datenschutz und Datensicherheit" lecture I had on uni in the '90s lying around somewhere in the attic.

But their meaning has changed and muddled so much over the years - probably not helped by the fact that "Sicherheit" is much closer to "security" in colloquial usage, and probably vice versa(?) - that they stopped being useful and used in this context.

I always thought it was from philosophy, Kant and the likes. Order, precision, detail (allegedly!).

Similarly for English and French, seen as practical and artsy, resepectively, due to say Hobbes/Smith and the likes of Baudelaire or Rimbaud.

Whether any of that makes any sense is a problem for the philologists, I suppose.

These things happen in any languages. English, for example, has "number" - which could mean cardinality of something (how many of something is here?), a number in the mathematical sense (real number, complex number, etc.), a digit (0/1/2/...), or a numeric identifier (room number, telephone number).

Also the infamous "free" bear vs. software.

I also work in Germany and the code-switching has nothing to do with the question of precision, but simply because English is the technical language for CS. Also, Germans apparently like everything American, so some of their own words which originally existed in German (and have exactly the same meaning as their English counterpart) have pretty much fallen out of use, cf. computer / Rechner.

It's not that German lacks precision per se but most of the jargon originated in the US or even England, and rather than coming up with German translations, it has become custom to use the original English. Which, frankly, makes everyday tasks like looking up documentation or debugging a lot easier.

Compare this to French where the Académie Française makes sure that you don't have to use these nasty English words! Yikes. And if there isn't a good French translation, they just make one up - my favorite example: the word "bug" (as in programming) has a made-up "French" alternative: "bogue". As far as I understand, no-one uses it, but it exists.

Also between painting and drawing. And between pumpkin and squash lol

Good reference to a higher level concept. Your linked article was a fun jumping off point.

What is “Simple Beans Factory Aware Aspect Instance Factory” supposed to actually mean?

That does not seem like a concept at all, let alone an actual German word. “Beans” is not even German, there is no German word spelled “Beans”.

How old were you when you first "learned" French? I've observed that kids/teenagers who aren't actually interested in learning the language retain just enough to pass their classes and then it all just drifts away. I was the same "learning" Spanish in high school in Texas.

Until I moved to a new country and _wanted_ to learn the language, I could barely remember how to ask where the toilet was. Now that I'm invested and interested, things are much more sticky.

I believe this is why much of the world has latched on to learning English. There is so much content available that people _want_ to consume, that it becomes a hobby they are actually interested in, rather than a chore. As more and more people learn English, it becomes a positive feedback loop of content creation that nearly the entire planet can participate in.

Or the tale about a speaker, being translated into German. He tells a joke, the English speakers laugh. He says "ok, and seriously now...", the German speakers finally hear the verb and start to laugh.

Ha! And don’t even get me started with the Scots and their whiskey. Bruichladdich, Pittyvaich, and Tè Bheag? Bunnahabhain Stiuireadair? Auchroisk??

I swear they only do this to mess with people.

So true. I always wondered why is Leicester pronounced as "lester" and not as "laichester".

The perks of coming late. Finnish did job properly with only one or two warts...

If you're counting sounds (i.e. phones), then most languages with those same 5 vowel letters have quite a lot because of numerous allophones.

English is criminal in that it has many vowel phonemes - British more so than American - and then on top of that its mapping of them to vowel letters is completely out of whack with both traditional Latin values of them, and how they're used in most other languages out there. I mean, cmon, <ea> for /i:/?

Still, with a sensible orthography, General American would only need something like 9-10 distinct vowel characters (for monophthongs; but diphthongs are best spelled as the obvious corresponding vowel-vowel or vowel-semivowel sequences anyway).

The question is: why hasn't the spelling been updated.

In my native German spelling, while there are irregularities, has been updated over times. Sometimes out of habit, sometimes by "order."

> First of all, it is not German, but Old English, which is not particularly more similar to German than it is to any other Germanic language e.g. Icelandic.

While English is changing relatively quickly, German isn't. Children today read original texts from the Gutenberg era in school without any trouble.

Where I grew up, we actually read some old english texts in school. It was hard, but certainly doable, using just our knowledge of modern German and English and the local dialects (north frisian and low german).

> I envy Chinese and Japanese;

Up until the point you have to read and write them.

Especially written Japanese, which is a giant mess of stuff they borrowed from the west, china as well as native stuff.

Measure words in Chinese are great. They provide so much descriptive capacity in such a short simple way. 一棍棒, 一把棒, 一根盪, 一條棒, all would translate to English as "a stick", but they convey different perspectives about what that stick is. I can appreciate the frustration with learning words that only have one specific measure word that only really describes it, but even then you can honestly get away with 個.

If you involve Swabian (which I could argue is more than just a dialect), everything goes out the window and you start again anyway.

And “weiblich” is the commonly used adjective instead of the Latin derived feminine.

> Wein, Weib und Gesang

This saying exists in English too: "wine, women and song".

Sorry to be picky, but "Wein, Weib und Gesang" is not neutral. It reduces "Weib" to the value of Wein and Gesang, something only needed for pleasure.

Fun fact: “Das Mädchen” (“little girl”, neuter) is diminutive form for “Die Mad” (“girl”, female).

All diminutives in German are neuter, for whatever reason. You could do the same for “Der Knabe” (“boy”) → “Das Knäbchen” (“little boy”).

Curiously, saying “Die Mad” would be as uncommon – at least nowadays – as saying “Das Knäbchen”.

A German guy once told me "Weizen und Weiber stößt man von unten"

Well, if it comes to that. German is not _really_ a single language. It’s a dialect continuum consisting of sometimes barely mutually intelligible variants. And yes, if you continue following that continuum, you get to the languages you mention.

A language is a dialect with an army and a fleet. As they used to say.

I can speak English and German which makes me able to somewhat understand written Dutch (especially if I know the context), but no chance when it's spoken.

Well, it's not a coincidence that the English word for the language of the Netherlands is the same the German state calls itself: "dutch" / "Deutsch".

A people and their language predated the concept of nation-states, but when the latter arrived obviously (geo-)political interests started to blur the facts.

So if you conflate the German state with Germans (I'd challenge that and view the German state as a continuation of the Prussian state), and you don't like the interests of the German state, it is predictable where you'll land on this issue.

Because of this, even if their national anthem does so, calling the Dutch Germans would infuriate them and rightly so, because it would imply justification to some for things like those happening between Russia and Ukraine right now.

I think in the end it is also a matter of "national" self-confidence. While Luxemburgish is virtually indistinguishable to the German ear from say the dialect of Cologne, Swiss-German is hardly understandable for anyone outside of Switzerland. Yet, the Swiss don't have an urge to re-label their dialect as a separate language. And the urge of the Dutch to re-lable themselves is lesser than that of Luxemburg because seemingly no one questions their identity.

A Dutch speaker read Afrikaans without too much effort; understanding spoken Afrikaans is a bit harder, but depending on the person it can be fine.

A Dutch speaker can't read or understand German. Some words are similar, but the same can be said about English. There are a number of differences in the grammar and alphabet.

Of course they're related languages; because I can speak English, German, and Dutch I can kind of read Swedish or Danish on account of being Germanic as well. But that doesn't make a "dialect with pretensions". We might as well say that all current Germanic languages are some sort of "dialect with pretensions" of some old Germanic language. But that doesn't really mean anything.

How is English then not a German dialect with pretensions? Is there a certain threshold in number of words that must originate from a different language family, or what's the logic here?

Yep. I find it easier to understand German verbally than Dutch. I struggle when Dutch people speak to me, the way they pronounce words are hard on my ears. German feels softer.

I did once tell one of them something quite like that.

I was assured with a smile that the feeling is mutual.

There's this fun video from Easy German, where native speakers (Austrians in this case) express their feeling of uneasiness when flirting in German. As in, the German language not being the sexiest.

https://www.youtube.com/watch?v=hEx0edmLdLo

Weaknesses can become strengths. Sometimes you want to have a quarrel. When French people quarrel they must rely on changes on pitch, cadence and volume because otherwise it sounds like they are ordering baguettes at the boulanger.

German has harsh sound. But in terms of quarrelling, I find that Korean sounds like someone is complaining all the time. But then again I have never learned Korean, so my impression would surely change.

I invite you to listen to Danes conversing before forming an opinion on German pronunciation.

> Schmetterling!

I feel the same about Afrikaans, ugly and harsh as hell :D

The French are so proud of their roots, they insist that for every French word you learn, you also learn its entire etymological development all the way back from Vulgar Latin.

One of the things that helped me improve German was Poetry Slam contests, they are still quite popular over here in many regions, you get poetry coupled with another German property, plenty enough sarcasm.

Lekker!

> French of course also have many original grammatical torture instruments.

For me it was when I had to realize that for the French, every number larger than what they can count with their fingers becomes a small algebra problem. quatre-vingt-dix-neuf ... four times twenty plus ten plus nine makes 99.

Laking the cases is Danish.

He actually learned German well enough to have appreciative audiences in Germany, but he also knows how to make amazing comedic essays on many topics. He did plenty about US-specific topics, and about French too, not just about German.

Which gives us Hitler memes where they audibly says German words that are very similar to their English counterparts, but the /funny/ subtitles is just a Beavis and Butthead level joke.

Doesn't work as good if one has ears.

Yeah, I vividly remember writing whole essays in school with only two sentences like that (sometimes over several pages of relative and temporal sentences and adjective chains), just as a raised middle finger to my German teacher who sleighted me once over the interpretation of some baroque poem.

Good times.

It's not so much the cases but the interaction with cases and gender. I found Finnish easier in some ways, despite the many cases. Because the case endings are always the same (modulo vowel harmony) so you can extract helpful information - something is inside something else say. In Czech a word ending with u has five different possible gender-case combinations (if I counted right) and that's not counting the distinction between short and long u.

I'm a Polish speaker and have met some Polish learners in my life. Often I have no better advice than "you choose the conjugation patern based on how does the word feel to you".

> the "ASCII" alternatives (ue, ae, oe, ss)

Isn't that called romanization? Similar to turning 刘慈欣 into Liu Cixin because you can't make the characters

I've found that germans take it for granted that it works this way, but I know of no other latin-script-based language that does romanization. Granted, I don't speak very many languages, but at least among the bigger ones like French, it's not like you write cafee (add an e because you dropped an accent), it's just cafe when you can't make the é. That's actually a terrible example, I just realized, because in german you totally use kaffee (yes yes, different word but same root). Let me try again with the word naive, coming from french naïve: you'd never write naieve. Or if you don't know how to make the ï in Dutch geïntegreerd, writing geintegreerd is understood by everyone whereas geientegreerd only leads to confusion. You could argue that it's because these ï don't have an "e" quality to them, but there is no other romanization taking place either for these, it's just dropped. Only Germans romanize to preserve the pronunciation-to-spelling mapping

On further research, this appears to be Microsoft's attempt to do something like a Compose key (https://en.wikipedia.org/wiki/Compose_key#Common_compose_com...) which I had forgotten about. In turn this is sort of emulating a "dead key" on mechanical typewriters (https://en.wikipedia.org/wiki/Dead_key), although I can't tell if German typewriters actually had a dead key for the umlaut or if they actually had additional keys for ä, ö, ü like modern German keyboards do.