Hacker News Reader

The CEFR is intended to measure the proficiency of foreign language learners and not measure the sophistication of native language speakers. Everyone that speaks English as a first language would pass a C2 level examination very easily.

The standard ways to measure text complexity in English for native speakers are the Flesch-Kincaid readability and grade level tests. These depend on the length of words and on the number of words in sentences so they're obviously not perfect, in particular they do not measure the use of rare vocabulary or references nor of difficult content.

I tested the OP and found: Flesch reading ease 53.2 (fairly difficult)

Gunning Fog 11 (hard to read)

Flesch-Kincaid Grade Level 10.1 grade

Coleman-Liau 10th grade

SMOG Index 9.6 grade

ARI 9.5 grade

Linsear Write Formula 11.4 grade

Where the references to grade are to American school grades. In other words this text can be read by average American 10th graders without difficulty.

I ran an article from the current New Yorker through as well (the one on Basquiat) and that came out as 11th grade reading level.

While it sounds like your measurements are more attuned to native speakers, I am not so sure that any high schooler could easily pass a C2 exam. Here is an example C2 practice exercise that doesn't seem all that far off from something you might see in the SAT: https://www.examenglish.com/cpe/cpe_reading_part1.htm

I doubt that the average native English speaker would pass a C2 level exam (even if you only consider the US and UK, as there are many other countries that speak English (for example India). While they probably won't have any problems with the grammar, C2 uses a lot of vocabularity that you won't be familiar with if you don't read much.

> Everyone that speaks English as a first language would pass a C2 level examination very easily.

This is almost exactly equivalent to saying that an average ten year old can read the New Yorker. A ten year old already has their native language down, the rules of grammar and syntax, the accent, the vocabulary, the capacity to discriminate between words with the same denotation but different connotations. There are ten year olds who can profitably read the New Yorker but very few. At least 10% of the adult population will have an absolute score on an IQ test similar to the average ten year old’s.

There are undoubtedly New Yorker articles that a ten year old could completely understand with sufficient exegesis but it’s the US’ premier literary journal and has been for decades. If they haven’t published articles with the depth and complexity of a Supreme Court judgment on some level I will be greatly surprised.

Most people are literally incapable of passing a law degree just as they are of learning calculus. You can write formal logic in English prose instead of using the symbols. The English prose version will be no more comprehensible to an average native English speaker than the one with mathematical symbols.

> exegesis

That's a ten dolla word if I ever saw one.

:)

>This is almost exactly equivalent to saying that an average ten year old can read the New Yorker.

It is not because it is fundamentally incorrect to say that the New Yorker is a C2 level text.

The CEFR is not intended as a scale to measure the complexity of written text for native speakers and/or readers of a language, it is intended as a four part (listening, reading, speaking, writing) assessment scale of the overall language abilities of people. At most, one could say that a text was consistent with what would be expected of a successful CEFR C2 candidate in the reading portion of the test.

One of the reasons that you cannot use the CEFR to characterise the proficiency of native speakers is that it assumes that you can understand texts of a similar complexity in your native language. Most such assessments are given to educated people, usually adults, so the texts reflect that.

Yes, there are people who are not highly proficient by the standards of a native speaker at reading and particularly writing in their native language. Who knows, I may be wrong to think that they would easily pass the written portion of a CEFR test, certainly they would smash the oral portions.

However the CEFR tests are fundamentally not designed to assess quality of education for native speakers. You could shoehorn them into that role but we already have other measures of language ability (in the US often calibrated to grade level) that are specifically designed to measure this.

There are a lot of problems with the way you approach this.

19% of Americans cannot read well enough to fill out a job application correctly.

The organization for economic cooperation and development found that 50% of U.S. adults can't read a book written at an 8th grade level, which is a statistic which obviously requires the estimate of "average American reading at a 7th or 8th grade level" be questioned heavily.

I've seen many estimates that the average high school student reads at a 6th grade level, and only 15% of the country is at an undergraduate level despite 30% of people having a degree.

I don't know where you get the idea that "grade level" relates to grades completed, but the average American is likely at a 5th grade reading level.

Well, I'll admit to not being an expert on the American school system with which I only had a few years of contact. If that really is the case, that grade level of reading ability and number of grades completed is not linked then that is certainly an indictment of the American system of education. I'm happy to amend my claim that 10th graders can read the New Yorker to say that by American academic standards, they should be able to but often are not.

It doesn't change the fact that the CEFR framework is not intended either to measure native proficiency nor to assess the difficulty of a written text.

I'm not disputing that the New Yorker is written at a higher level than the average American cares to read, just disputing the use of CEFR framework to assess the text.

> The CEFR is intended to measure the proficiency of foreign language learners and not measure the sophistication of native language speakers.

You’re right about this, but these are the systems I know, and they are arguably more useful in this context that most of the grade-level models you listed (very abstract, and most folks don’t really understand the level they are based on).

> Everyone that speaks English as a first language would pass a C2 level examination very easily.

C2 is the equivalent of a highly educated native speaker. This does not describe most Americans, even ones with seemingly appropriate credentials.

A discussion I’ve heard about native speakers of English expected to pass a C1 test in a foreign language: “Can they pass C1 in English?” The answer was not always so clear.

> In other words this text can be read by average American 10th graders without difficulty.

These are more like ideal goals for a high-performing Xth grader. I have worked with the vocab lists that textbooks writers are supposed to use — the expectations are very high.

Anyway, thank you for your response.

> It will sometimes cause less proficient readers to take away substantial misunderstandings about the contents of the article.

The most prolific example of the reading level mismatches causing trouble is Paul Krugman's August 2002 column: Dubya's Double Dip? https://www.nytimes.com/2002/08/02/opinion/dubya-s-double-di...?

The famous sentence from the article:

> To fight this recession the Fed needs more than a snapback; it needs soaring household spending to offset moribund business investment. And to do that, as Paul McCulley of Pimco put it, Alan Greenspan needs to create a housing bubble to replace the Nasdaq bubble.

If you read the whole article and think that Krugman says this housing bubble is a good idea your reading level can't be that good. Too many people did read the whole article and failed to grasp the tone and nuance. Even Arnold Kling had to point out this to his readers http://www.econlib.org/archives/2009/06/defending_what.html

Maybe Krugman’s writing is just actually very unclear. “Difficult” writing is to be avoided, not praised.

One of the marks of a good writer is how they uncomplicate the writing without watering it down. Krugman is really good writer in that regard. His writing skill was probably a big reason why he became regular columnist in NYT.

That kind of writing can still be inaccessible to lower reading comprehension levels. If you are uncertain of his writing skills, I don't know what to say.

This is only true if you aim to reach as wide an audience as possible. If you are writing for a specialist audience making the reader work to connect the dots themselves has an extremely long history.

https://econjwatch.org/articles/a-beginner-s-guide-to-esoter...

> This article is a republication, by permission, of a chapter of the author’s book Philosophy Between the Lines: The Lost History of Esoteric Writing (University of Chicago Press, 2014). The piece provides a beginner’s guide to techniques and devices used in esoteric writing. Among the techniques and devices described are the following: dissembling the true message (sometimes by presenting it as from a disputant, beggar, or buffoon, sometimes by arguing against it in ways that enhance awareness of its truth); dissembling the true target (exoterically speaking of Y when the real target is some other thing Z); developing a compelling argument and then taking it back; textual incongruity (for example, departing from a declared plan); conspicuous inconsistency or self-contradiction; the commission of errors that the author’s demonstrated competence and mastery would not allow (for example, altering a quotation in a significant way); dispersal (dispersing argumentation for a tacit viewpoint throughout the text); expressing very striking or intense thoughts in an oblique or ancillary fashion, such as in a meandering digression or in the notes; meaningful silence or conspicuous omission (as when the text creates expectations of coming to something that then remains unaddressed or unstated); alluding subtly to the writings or opinions of a significant figure; and placing thoughts of particular significance in middle of the text or in the exact center of a list or sequence.

>has an extremely long history.

That's really not a meaningful argument as to its merit. In fact, it's so famously not a meaningful argument that it has its own fancy name: argumentum ad antiquitatem.

I don't believe they were making a case that it has merit, only that it a long history and is a specific writing conceit for its audience...which I suppose gives it merit as being appropriate for its audience.

Not to mention that the typical New Yorker article is looong. Way longer than the typical random internet comment. This puts some demands on the reader as well. :-)

>C2

I wonder if there is a similar scale for cartoons, and if (inscrutable) New Yorker cartoons are at D1 or higher.

1. I don’t think that there is a D1 on the scale.

2. A scale for cartoons would be interesting. The top level would probably require something like awareness of current events, knowledge of one or more complex/academic theoretical concepts, and deep cultural embeddedness and awareness of cultural trends.

"On a scale of Marmaduke to the New Yorker, how complex is this comic strip"

From all the US satirical cartoons I have seem they seem rather crude and tame compared to the UK and still rely rather old tropes that always seem to require labelling "oh look its the cat with $ signs for eyes and labelled "wall street"

whereas in the UK a cartoon might make a reference to a cartoon from 50-60-70 years ago and you would be expected to understand the reference with out labelling

Do you have examples of these sorts of cartoons? I've never seen one so I can't imagine what they're like.

So here's an example of a Steve Bell editorial cartoon. He's a leftwinger, in the fairly angry and vicious tradition of British satire, but this is a fairly restrained one:

https://www.theguardian.com/commentisfree/picture/2019/jun/0...

It's a comment on the pageantry around the 75th anniversary of D-Day, and directly refers to a famous cartoon from WW2, which was published on VE Day by the (also socialist) cartoonist Philip Zec:

http://www.cartooncounty.com/serendipity2/uploads/Zec2site.J...

The whole genre isn't self-referential, but it's not unusual for leader cartoons to refer to other cartoons, famous paintings, television, or other things.

Anything by Ben Garrison https://grrrgraphics.com/cartoons

New Yorker cartoons are not political cartoons. They are general humour, albeit of a sort that might strike most readers as highbrow.

Slight mistake that I can’t edit — New Yorker articles can range from C1-C2... maybe even down to B2 at times.

Regardless, all of the main points still hold.

Do you have some examples here?