Inferring New Vocabulary from Context in Novels - Your Experience?

[uncertaingoblin]

Hi folks, I'm very interested in what you have to say about vocabulary acquisition purely from context. Have you tried to learn new vocabulary in this way? Although I imagine it is not as efficient as looking up unknown words, I find this concept incredibly interesting. If you have tried it:

Did you find that you learned some new vocabulary, or did you find it didn't do anything?
Did you enjoy it more or less than when you looked up unknown words while reading?
If you think it can work, what level of comprehension do you think is necessary?

Thank you for sharing your thoughts and experiences on this topic.

[desafiar]

From having read archives of the forum, I suspect I'm in the minority, but don't enjoy inferring meaning by passing over unknown words for the purpose of extensive reading.

If I don't know a word or have a low confidence in my understanding of it, I look it up every time. While reading progress is slightly slower, I find it more satisfying overall, and I'm not convinced that in the long run I'll progress faster by inferring. For the small hit to speed, I gain a better understanding, and I think that helps me.

I do like to minimize time cost of lookups as much as possible. For that reason, I've put aside my Kindle device and now read on my Macbook. Doing so gives me the advantage of faster dictionary access and sentence translations. They are two clicks away. Given the number of lookups within a reading session, the increased speed over that of the Kindle is significant. This method also makes pasting sentences into another document for later analysis or inclusion into Anki fast and easy.

I also adhere to the principle of N+1. And I'll continue doing that to the extent it's possible. Right now that means completing all the graded books from Paco Ardit. After that, I'll read a handful of books from a single author, and then move to another author in the same genre. Always moving forward N+1.

I do believe the tortoise can win sometimes, and enjoy the scenery along the way.

[Yunus39]

I'm looking forward to reading the responses!

It really depends on your personality and how comprehensible the input is. Some people have a higher tolerance for ambiguity than others.

If you have a 95+% comprehension of what you are reading, you shouldn't struggle too much. The less comprehensible it is on the whole, the less context you have to get meaning. Bilingual texts can be really helpful at this point. If you have already read the book in your native language, it is going to be much easier to infer the meaning of new words in context.

There are different methods you can use. I like to underline every word I don't know whether I look it up or not. If you do that, you will definitely start to notice words that show up multiple times in the same chapter. You should probably look those up if you cannot get them from context.

Distilled from Olly Richards:

Pick a book that you can read and understand the broad story. Read a chapter once without looking up a single word. Read the chapter again, highlighting any words you don’t know. If there are any words that show up five times or more in the chapter, look them up. Read the chapter again (3rd time).

[sporedandroid]

I find reading a lot more enjoyable when I can look up words. Hebrew words can also look incredibly similar to each other. So I sometimes think I know a word when I actually don’t.

[StringerBell]

Did you find that you learned some new vocabulary, or did you find it didn't do anything?

I found that in the beginning/early intermediate phase, inferring high frequency words in simpler texts worked ok, especially if there weren't too many unknowns. My preferred method is to guess an unknown word's meaning and then look it up to confirm if I'm right. Sometmes I'm right and sometimes... I'm really off. I personally think that relying too heavily on inferring unknown words is not an efficient strategy for improving vocabulary and I just don't like it (and don't even do it in my native language) but that's just my opinion. I tend to only rely on inferring a word's meaning when it looks like a cognate or I've come across it multiple times AND I can very clearly tell from the context what it means (otherwise there's a risk it's a false friend or it doesn't mean what I assume it means because it's too easy to "force" the misunderstanding to work with the context).

Did you enjoy it more or less than when you looked up unknown words while reading?

I enjoy what I'm reading significantly more when I can look up (or in some other way) get confirmation of unknown words/phrases. I find that if I rely on guessing too much I end up with a hazy understanding of what I'm actually reading, which too frequently leads to misunderstanding or losing details. When I read, I don't want to just have a vague sense of what's happening and I don't enjoy ambiguity. I don't mind guessing a word once in a blue moon but I don't like to rely on it regularly as a strategy.

On this forum you'll find everything from people who look up almost nothing to people who look up almost everything with varying results all around. My opinion on the topic is evolving as I experiment with various strategies. Based on the early (and significant) success I'm having with my current experiment (repeated rereading with dictionary look-ups during only the first reading), I sincerely regret the time I wasted on extensive reading (where I either guessed a meaning and moved on or quickly looked up a definition on my kindle and moved on). I remember few of the words I encountered using that method. I can confirm this by looking at the unknown words I highlighted in my kindle during my extensive reading phase. Not only do I have no idea what the vast majority of them mean, but I can see that I highlighted multiple forms of the same word, which tells me that I wasn't recognizing words repeating even within the same book.

I'm sure I did learn some words from context by guessing but the number that I learned and remember compared to the time I spent reading (which at this point is 17 unique novels, ~7,000 pages) is a pretty poor return on investment. In the past, the best results I got with vocabulary improvement came from intensive reading and making & studying flashcards of the unknown words I came across in books.

However, using my current method I'm remembering close to 100% of all new words and phrases encountered during rereading and I'm not even studying any of them with flashcards. I can also recall many of these new words after I'm done reading and use them, myself, when I'm speaking. I'm recognizing them when I see them in other sources, like YT videos. None of these things were happening with the unknown words I encountered when I was doing extensive reading in the past. As soon as I was done reading the page (or the sentence) that unknown word was out of my mind as if I'd never seen it at all.

3-5 unknown words per page (98.8%-98% comprehension) in a 300 page novel is 900-1,500 unknown words per book. When I was extensive reading, I was going through 1 book per week. The more material, the better, right? Well, not for me. That many unknowns was overwhelming and my brain dealt with it by forgetting about all of them. Kind of like that famous scene from I Love Lucy in the chocolate factory... when the chocolates (unknown words) come at me slowly I can deal with them but when there are 100-200 of them per reading session it's a disaster. Extensive reading was endlessly frustrating and the promised "improvement" that was supposed to magically happen somehow didn't for me. Maybe I just suck or maybe it would have eventually worked if I could have stuck to it for 100 actual novels but what I'm doing now is working so much better that I can't really see a justification to go back to it.

Now I can actually see immediate improvement in my reading skills within the same reading session. I'm documenting what I'm doing in my log if you want to read about it or ask questions (or make suggestions). What I'm doing with reading is essentially the same strategy that I used for improving my listening comprehension. It worked extremely well for listening and I don't know why it didn't occur to me to do it with reading sooner.

I expect someone will show up shortly to argue with me but my experience is that extensive reading where I rely heavily on context clues to guess at meanings isn't an efficient way for me to learn or remember new (lower frequency) vocabulary. Clearly it works for some others so I suspect there is no one answer that works for everyone.

[uncertaingoblin]

Did you find that you learned some new vocabulary, or did you find it didn't do anything?

I found that in the beginning/early intermediate phase, inferring high frequency words in simpler texts worked ok, especially if there weren't too many unknowns. My preferred method is to guess an unknown word's meaning and then look it up to confirm if I'm right. Sometmes I'm right and sometimes... I'm really off. I personally think that relying too heavily on inferring unknown words is not an efficient strategy for improving vocabulary and I just don't like it (and don't even do it in my native language) but that's just my opinion. I tend to only rely on inferring a word's meaning when it looks like a cognate or I've come across it multiple times AND I can very clearly tell from the context what it means (otherwise there's a risk it's a false friend or it doesn't mean what I assume it means because it's too easy to "force" the misunderstanding to work with the context).

Did you enjoy it more or less than when you looked up unknown words while reading?

I enjoy what I'm reading significantly more when I can look up (or in some other way) get confirmation of unknown words/phrases. I find that if I rely on guessing too much I end up with a hazy understanding of what I'm actually reading, which too frequently leads to misunderstanding or losing details. When I read, I don't want to just have a vague sense of what's happening and I don't enjoy ambiguity. I don't mind guessing a word once in a blue moon but I don't like to rely on it regularly as a strategy.

On this forum you'll find everything from people who look up almost nothing to people who look up almost everything with varying results all around. My opinion on the topic is evolving as I experiment with various strategies. Based on the early (and significant) success I'm having with my current experiment (repeated rereading with dictionary look-ups during only the first reading), I sincerely regret the time I wasted on extensive reading (where I either guessed a meaning and moved on or quickly looked up a definition on my kindle and moved on). I remember few of the words I encountered using that method. I can confirm this by looking at the unknown words I highlighted in my kindle during my extensive reading phase. Not only do I have no idea what the vast majority of them mean, but I can see that I highlighted multiple forms of the same word, which tells me that I wasn't recognizing words repeating even within the same book.

I'm sure I did learn some words from context by guessing but the number that I learned and remember compared to the time I spent reading (which at this point is 17 unique novels, ~7,000 pages) is a pretty poor return on investment. In the past, the best results I got with vocabulary improvement came from intensive reading and making & studying flashcards of the unknown words I came across in books.

However, using my current method I'm remembering close to 100% of all new words and phrases encountered during rereading and I'm not even studying any of them with flashcards. I can also recall many of these new words after I'm done reading and use them, myself, when I'm speaking. I'm recognizing them when I see them in other sources, like YT videos. None of these things were happening with the unknown words I encountered when I was doing extensive reading in the past. As soon as I was done reading the page (or the sentence) that unknown word was out of my mind as if I'd never seen it at all.

3-5 unknown words per page (98.8%-98% comprehension) in a 300 page novel is 900-1,500 unknown words per book. When I was extensive reading, I was going through 1 book per week. The more material, the better, right? Well, not for me. That many unknowns was overwhelming and my brain dealt with it by forgetting about all of them. Kind of like that famous scene from I Love Lucy in the chocolate factory... when the chocolates (unknown words) come at me slowly I can deal with them but when there are 100-200 of them per reading session it's a disaster. Extensive reading was endlessly frustrating and the promised "improvement" that was supposed to magically happen somehow didn't for me. Maybe I just suck or maybe it would have eventually worked if I could have stuck to it for 100 actual novels but what I'm doing now is working so much better that I can't really see a justification to go back to it.

Now I can actually see immediate improvement in my reading skills within the same reading session. I'm documenting what I'm doing in my log if you want to read about it or ask questions (or make suggestions). What I'm doing with reading is essentially the same strategy that I used for improving my listening comprehension. It worked extremely well for listening and I don't know why it didn't occur to me to do it with reading sooner.

I expect someone will show up shortly to argue with me but my experience is that extensive reading where I rely heavily on context clues to guess at meanings isn't an efficient way for me to learn or remember new (lower frequency) vocabulary. Clearly it works for some others so I suspect there is no one answer that works for everyone.

I think this is a very informative post, thank you. It highlights that inferring meaning works for some but not for others and even that types of lookup methods don't work for everyone. One thing I think it does make clear is that if a non-lookup/inferrance only method works, it requires a massive amount of content. Perhaps the required amount of content to make that work speaks to why people can succeed in this method as learners of English but not so much with other languages.

[BeaP]

I enjoy reading without looking up words a lot because I mostly read for pleasure. Maybe I learn some new words this way but I'm not sure. My impression is that I don't acquire anything new, only improve my reading skills (speed, understanding of nuances).

Sometimes I look up recurring words or those that seem important. If I don't write them in my notebook and repeat them later, I only remember a small fraction of them.

However, if I also do other activities on the same level, things add up, and reading becomes a much more efficient tool. For example, when I study from a textbook, I regularly meet the new words during my reading, even on the C levels, where everything is supposed to be so rare that this shouldn't happen. Also, discussing the book with someone or reading reviews helps a lot, because the key words are usually highlighted and it's easier to infer them in these short and focused contexts.

[einzelne]

I find that if I rely on guessing too much I end up with a hazy understanding of what I'm actually reading, which too frequently leads to misunderstanding or losing details.

I cannot locate it now but I remember a SLA study which basically demonstrates that people suck at inferring the meaning from context. Something to keep in mind when approach reading in L2.

[Iversen]

I hardly ever read novels these days, but I do read other types of text, including popular science articles and travelogues - and the experience can't be very different. And I am in the paradoxical situation that I don't like not knowing the meaning of the words I see, but on the other hand I know from experience that I'm fairly good at hammering through a text in spite of unclear passages - I use that for extensive reading in languages which I still don't know very well or in dialects or related languages. For instance recently I used a translation into Frisian (Frysk) in one of my bilingual texts even though I don't even own a dictionary in the language, and it went smoothly. But basically I prefer knowing the majority of the words in my study texts, and I don't even try to memorize words I meet unless I am reasonably sure what they mean - no reason to commit false guesses to memory. And this is also the case when I listen: I simply don't remember unknown words from things I hear - I need to see them and if necessary check them in a dictionary before they stick in my mind.

I once made a lecture in Novi Sad about this topic, and I had found one article where 'ordinary' readers in an experimental setting were subjected to supposedly unknown words in a text (as far as I remember the words were constructed for the purpose). The special thing about this article was that it used both multiple choice and 'free' questioning to check whether the testees had remembered anything, and the result was that there only was a measurable impact if controlled through multiple choice (where only the basic recognition is tested), and the result was not impressive at all. With 'free' questioning (which tests active recall) the success rate fell to almost zero. This doesn't mean that you can't learn from reading, but the method seems to be inefficient and probably depends on the degree to which you can sustain an alert curiosity towards the words in the text - and not just the general meaning or gist. Paradoxically you may be better off trying to learn from a text which you only partially understand because you then have to concentrate on the words. In an wellknown language you are more likely to skip the few remaining unknown terms.

[einzelne]

I remember few of the words I encountered using that method. I can confirm this by looking at the unknown words I highlighted in my kindle during my extensive reading phase. Not only do I have no idea what the vast majority of them mean, but I can see that I highlighted multiple forms of the same word, which tells me that I wasn't recognizing words repeating even within the same book.

I had exactly the same experience! That's whyI 'm trying to practice deliberate vocab review time and then.