I use Anki and the MOST consistent reason a word lands in the leech pile is not because I can’t remember what it means, but because I consistently think it means something different than what it means.
So for, instance, I keep thinking French costaud means, “rib or side”, probably because I associate it with *costal" from anatomy class or costilla from Spanish.
What I do in this instance is make a Match Pairs card with:
- costaud matching “strong, burly”
and - côte matching “rib.”
This way I practice differentiating them and simultaneously learn the correct word for rib.
But I do worry that I’m unintentionally strengthening the connection by having them always together like this.
What do you guy do for these situations?
Edit: You may notice that a properly formatted card would include le/la to indicate the gender of the noun, which costaud, as an adjective, would not have but I am inconsistent and also don’t like depending on little tricks that only work in index cards, so for the matching card I do not write “la côte” because that way I could guess the card based on part of speech instead of definition.
I find that digging into the words help. In your case, you’re mixing them up because they are indeed related! Somebody who’s costaud has strong côtes (according to the CNRTL, they are not 100% sure either).
In my case with Chinese words I learnt to write them so I’d distinguish 我找钱 much more clearly, but I still get the odd mistake like when I confused 午(noon) next to a horse for 牛 (cow)… I was very confused as to why the horse was labeled as cow. But yeah, the general idea is to have more data from more senses to sort the words better in my brain.
Their shape, their sound, the way to write them (physical movements), a picture of what they represent, an anecdote tied to the word, or in this case to learning the word, etc.
Ah, yes, etymology got me a long way with my native language, but I never dive into it for my foreign languages, partially because I’m just in a hurry…
They’re also usually located in native language dictionaries, the reading of which make me break out in hives.
Speaking of which:
I didn´t know about the CNTR but here is the etymologie of two related (?) words from that site and I can’t make sense of it, honestly. I’m used to an etymology tracing the root word from latin, (then often what it turned to in French), to what that word turned into in English; yet, I see no mention of any root words at all in these entries:
bête
- Subst. ca 1100 beste « animal [terme générique] » (Roland, éd. Bédier, 1597); 1160-70 au fig. [en parlant d’un homme] « être stupide » (Béroul, Tristan, 1309 dans T.-L.);
- adj. ca 1220 id. « stupide [en parlant d’un homme] » (G. de Coincy, Mir. Vierge, 321, 294, ibid.), seulement en a.fr. (T.-L.); repris au xviiie s. 1763 (Diderot, Salon de 1763 cité par Greimas dans Fr. mod., t. 23, p. 138). Empr. au lat. bēstĭa qui, d’une part s’est transmis par voie pop. sous la forme bīstia (v. biche, a.fr. bisse), d’autre part s’est maintenu sous la forme class. (bēstĭa) et s’est transmis par voie d’empr. par l’intermédiaire d’une forme *bestie, v. Fouché, Phonét., 417. L’hyp. d’une formation pop. à partir de *bĕsta, Dauzat 1968, EWFS2 (forme non attestée de manière sûre — v. TLL, s.v. bēstĭa, 1935, 34-37 —, mais déduite du dimin. bēstŭla, Fortunat, Vita Martini, 3, 341, var., ibid., s.v. bestiola, 1941, 27) fait difficulté parce qu’elle fait appel à une 2e var. de bēstia (bīstia postulé par a.fr. bisse est bien attesté).
embêter
- 1794, janv. « causer une vive contrariété à quelqu’un » ceux qui les ont « embetté » (Arch. mun. de Passy, Délibération ds Brunot, t. 10, p. 225, note 6); 1794, 20 juin Frederic-Guillaume est… embêté par cette ridicule secte (Bull. Conv. Nat., ibid.) ; 1831 embêté part. passé adj. (Sue, Atar Gull, p. 22); 1842 embêtant part. prés. adj. (Flaub., Corresp., p. 91); 1876 embêtant part. prés. subst. (Goncourt, Ch. Demailly, p. 113 : l’embêtant est que ça revient toujours);
- 1820, 19 avr. « causer de l’ennui » (Stendhal, Corresp., t. 2, p. 188 : M. Taglioni et sa femme nous ont embêté ici d’un ballet à la française); 1840 embêtant part. prés. adj. (Matelots et Matelottes, ii, in Répertoire dram., IV ds Quem. Fichier); 1842 embêté part. passé adj. (Flaub., Corresp., p. 93); 1874 pronom. (Id., ibid., p. 135). Dér. de bête1 ; préf. en-; dés. -er. https://www.cnrtl.fr/etymologie/embêter/verbe
Haha! Yeah these are hard to parse, even for us. It’s full of abbreviations, for a start. But they do tell you the story of the words as best as is available, and do tell you clearly when a link is a hypothesis, a guess, etc. With references.
So bête comes from beste in old french (ca 1100 means circa 1100, ie, around the year 1100). A.fr. is an abbreviation for ancien français. It’s interesting because similarly for the adjective version, it starts around 1200 (so I’m guessing we have direct sources with the word appearing) and then in the 18th century dictionary, we start discussing the path from latin towards French. By successive borrowing (Empr. meaning emprunté, ie borrowed).It’s not always super clear and concise because, well, language isn’t ;)
Yeah, these are annoying!
The main thing I’ve changed that helps with that, is I’ve started putting the example sentence on the front of the card.
I like it for a few reasons. The point at hand - disambiguation. A lot of times in context, it’s obvious which answer I should remember. But also, then I read the example sentences more often. When they’re on the back, I tend to skip over them if I’m powering through several reviews. In the case of Japanese, this also gives me an opportunity to try and read the whole sentence w/o furigana.
Downside is that the example sentence can make it too obvious. This is actually totally fine IMO. The whole point of the Anki card is to prepare you to actually use language. If the word is clear in context, great, that’s the whole point. It’s also not too hard to ignore the sentence if you are in that mood.
But then, sometimes I still mess up similar words even in context. Rather than let it leech, I add a caution note.
Edit: re-reading, I realized that my mind went to a slightly different kind of disambiguation. Still, I think the answers apply all the same.
If meaning is clear from the context of a single sentence, it probably doesn’t even belong in the deck. That’s kind of my litmus test for harvesting–if I have a pretty good idea what the word is, I move on and soon find a word I have no idea about.
Don’t use Anki for this. Instead learn meanings through reading (including reading the dictionary in the target language).
I see Anki as the bridge between ignorance and immersion. You should only study the most common meaning to get started in the language. But you must always review the word in the dictionary.
At A1 or even A2 you aren’t strong enough to read a native dictionary however. But at B1+ (well, I’m at A2+) you will suddenly notice that the native dictionary isn’t so bad anymore
At A2+ I can read the example sentences and different meanings. I don’t understand it all but I can learn from dictionaries now…
A2+ meaning I completed A2 months ago but am still very far away from B1.
Anki is just good enough for seeding your understanding. The hard work happens at the immersion step. Don’t try to make Anki do something it’s bad at.
Use a dictionary. Seriously. If you aren’t strong enough for a native dictionary use a bi-lingual dictionary instead.
Gotta disagree with you on native language dictionaries. Even in my best language, Spanish, where I am conversationally fluent, dictionary definitions are often impenetrable.
For instance: I could make a card with sapo on one side and the dictionary definition anfibio anuro de cuerpo rechoncho y robusto, ojos saltones, extremidades cortas y piel de aspecto verrugoso on the other.
But words I don’t know in this definition are: anuro, rechoncho, I’m not sure what robusto means here, or verrugoso.
The definition of anuro is: Que carece de cola.
I don’t know carecer.
Carecer means lacking. Ok. So anuro means lacking a tail.
Rechoncho defines as grueso y de poca altura ok–so thick and short.
verrugoso means que tiene verrugas ok and verruga means excrecencia cutánea por lo general redonda.
Exrecencia means Protuberancia, generalmente carnosa, que se produce en animales y plantas, alterando su textura y superficie natural.
but…
**Or **
I could just write frog.
I’m guessing you’re a reader, but my interest in languages is mainly in conversation, which has a significantly smaller vocabulary; I make do with word reference bilingual.
I agree with you on everything else. The true nuance of a word is learned, as you say, from context, so knowing the root/literal meaning of a word is the core of what your anki study can give you. I used to learn all the different definitions, but usually there is a core meaning and the other meanings are merely metaphorical extensions, and it’s good to practice staying flexible.
For instance: I could make a card with sapo on one side and the dictionary definition anfibio anuro de cuerpo rechoncho y robusto, ojos saltones, extremidades cortas y piel de aspecto verrugoso on the other.
Oh no. That’s not what I mean at all.
Again, I CANT read a dictionary yet. I’m A2 lol. But it’s still useful!!
Read what you can. At A2 level you should have already mastered the words noun, verb, adverb, adjective, preposition (etc. etc). How many categories is the word in?
How many definitions? Is this a flexible word with 50 definitions or is it a simple word with 3 related definitions under one subheading?
What’s the frequency count of the word? Are there any synonyms or antonyms that you recognize?
Are there any “related” prepositions with the word? In German, it’s “helfen bei” (which is different than English “help with”), the native level dictionary makes it clear which preposition is closely associated to the word.
Even though I can’t read definitions, there’s still a TON of information here that I learn from.
I NEVER make the Anki card from a native dictionary.
No, I make very simple Anki cards. I then spend lots of time in the dictionary (native and bilingual) to learn more about the word. Anki remains a simple one definition per word flashcard.
You never answered my question about reading. I think your approach makes a lot of sense for someone who really needs to boost their vocabulary to get through literature. When I was learning English, my native language, we’d also learn etymology (the root and derivation), and that really helps me understand unfamiliar vocabulary, but it was a huge investment.
I mean, I plan to learn everything I can.
There aren’t much German speakers here in the USA. So I can’t get that much conversation practice. But I do want to be conversational, eventually. I regularly go to a club for that.
But again, I’m A2. I’m seeking the conversational level but I’m basically a beginner in the great scheme of things.
I agree with you that conversations I’ve had seem to have a lower count of vocabulary. But all of these nuanced meanings of words used in practice are different between the textbook and real life conversations.
Ex: I’m glad I looked up the word halten (hold, also “Halt”) in a real dictionary. It really shows how flexible the word would be used in a conversation.
I agree that there are some words better for conversation, and other words better for reading. But in all cases, studying the dictionary and being aware of the variety of uses (even if it’s nearly unreadable) is useful to me.
Yeah, we “learn” languages when we don’t have the opportunity to “use” languages…
I got really gungho about vocabulary when I was learning Spanish and I found all these delightful words I learned reading books. But when I went to use them with my friends, they’d be confused and correct me. So I decided I would avoid literature from then on, but read books written more accessibly (like right now I’m reading a french autobiography that is written very conversationally).
How has your experience been with language exchanges? I gave up on it. Too many lonely hearts of both genders.
Hmmm, German speakers here in USA seem to be mostly retired folk. I wouldn’t say lonely but they are all like +20 or +30 years older than me.
I’d be nice to have someone closer to my age but I welcome any practice.
Usually, mnemonics are my nuclear option if it gets to this point. I guess one could say I have a sliding scale with my vocab, where I step the word up if it’s problematic:
- base word
- base word with sentence
- base word with sentence and meaning note
- base word with sentence, meaning note and/or mnemonic
Here’s an example. A while ago, for some reason I constantly mixed up 転ぶ (to fall down/over, pronounced korobu) and 運ぶ. (to carry/transport, pronounced hakobu). hakobu is a very high frequency word, so even seeing/hearing the word wasn’t enough. I initially tried a visual mnemonic with the kanji, which didn’t stick (and French doesn’t have, of course). Eventually I connected the word hako, meaning “box,” to hakobu.
hako wo hakobu. Carry a box.
It’s been much stickier since, although I still sometimes find myself sometimes briefly pulling up the binary choice and the mnemonic before selecting correctly. The association’s still there, but it’s faded a lot since I’ve put in more time with the language.
A lot of children songs have cartoons or dances associated with them.
The traditional German song Himpelchen und Pimpelchen has a little finger dance that represents the two characters climbing up a mountain for example (or for us Americans, basically a more German / different version of Itsy Bitsy Spider).
I got my start in German with Tschu Tschu wa, a song for maybe pre-schoolers thats like hands out, on the shoulders (etc. etc.).
So I guess what I’m saying is… I don’t really have mnemonics per se. I have dance moves or finger movements that represent different prepositions or verbs.
I have to imagine a lot of pre-school level (3 to 5 year old) songs have these dance moves. Like Baby Shark in USA had little dance things to represent the different sharks.
I’m old enough that I’m willing to learn like a 5 year old again. It’s an effective means of memorization!!
hako wo hakobu. Carry a box.
Nice!
Context sentences are good too.
NL --> TL is another option, I think, because the association might not be as strong in that direction.
