I might as well let it be known that I'm currently a graduate student at Technion, the Israel Institute of Technology.
The funny part is that after leaving undergrad, I had never imagined the thing I would dislike most about grad-school would be the coursework. It's not horrific and I'm certainly not incapable of it. It's just run entirely according to the Engineering School Mentality, which results in fairly bad teaching methods being used and a whole lot of antagonism between faculty and undergraduates, in which us grad students basically have to "pick a side, we're at war".
What is the Engineering School Mentality? "What matters is the grade, specifically on the final exam." It's a fairly different philosophy from what I saw in undergrad at UMass Amherst, and I'm somewhat disturbed by its... outgrowth into a full-blown teaching method. Its full expression is best captured in a veritable proverb I heard from one professor: "Students will do anything for a high grade on the exam; in extreme cases they'll even learn."
Let me explain. Technion does not handle coursework in the fashion of an "American-style" university, with 50-90 minute lectures several times per week per course and a steady stream of homework assignments, quizzes, in-class participation requirements, and mid-sized exams. Instead, it books one 3-hour lecture per course per week, gives a comparatively small amount of homework, requires only a moderate amount of in-class participation, and then puts everything riding on final exams and semester projects.
Exams period here is two months: one month to attempt your exams the first time, and the second month to attempt them again because you didn't do well enough. Of course, systems like this have "worked" for decades in engineering schools around the world. They certainly produce results: when "only the strong survive", you at least guarantee those who survive are probably pretty strong
What makes this a subject worth writing about at all is the bizarre contradiction of applying a teaching method designed for mechanical engineering to, of all things, a natural language class. As a foreign-born graduate student, I have to take and pass the course "Reduced Hebrew", which despite its name, is only given to those who already possess a moderate level of conversational spoken Hebrew. And it is ridiculous.
There seems to be little logical ordering to the curriculum, and the vocabulary and grammar taught are exclusively those of academia. For example, we learned words like "journal article" (ma'amar ktav-et) before we learned participles of common verbs. Worse, the teacher forbids herself from ever switching out of Hebrew to explain anything, resulting in our having to try to learn adult, academic Hebrew based on the pidgin dialect of small children and new immigrants. It also gives us continual utterances of "exactly the same thing" (bidyuk oto davar) or "X is like Y" (x zeh kmo y) for pairs of words or phrases that, when you try to actually use them outside class, mean something subtly but significantly different.
Worst of all, the course consists of zero free-form speech or writing. Question: what kind of language course enters its fourth month without one lengthy writing assignment or in-class unstructured conversation? Answer: a language course invented by an engineering school. The complete and total emphasis is on precisely memorizing grammatical forms and vocabulary sets for accurate regurgitation in exercises and exams. Any attempt to actually speak the language is not merely unrequested but almost punished: the most common response is, "We don't say it thus" (Anachnu lo omrim et zeh kakhah.). Colloquialisms are taught rarely, if ever, and when mentioned are discouraged with a disturbingly satisfied "It could be, but this is the prettier Hebrew" (Yekhol lihiyot, aval zeh ha'Ivrit yoter yafah.). The native idioms and metaphors of Hebrew, or matters of local culture, have not come up, ever. Speaking becomes more of an opportunity to "lose points" and interrupt the flow of the curriculum than to communicate one's thoughts.
The result is that I can feel my brain malfunctioning: my ability to read, hear, and understand Literary Hebrew (as opposed to the less Aramaic, more Arabic-and-English "street Hebrew" spoken colloquially by much of the population) has grown quite significantly in the time I've taken this course... while my ability to speak or write the language on my own has only scarcely improved.
And what's the point of such a course as "Reduced Hebrew"? It's what you might have guessed: to pass an exam that fulfills a university requirement. This seems like the textbook example of premature optimization, or optimization on the wrong criterion: "We have a metric, and we optimize for that metric. It's up to thestudents children from there, and the strong will survive."
I mention all this to contrast it with the ulpan system of language learning, the normal mode of immigrant language-absorption in Israel, which chiefly emphasizes free-form conversation and writing. And it works! A diligent ulpan-goer with even the slightest ability for languages will usually significantly improve their conversation and writing in even the five months of government-provided absorption ulpan, speeding up the absorption process and enabling new immigrants to join society. There are no exams until the very end, and the final exams consist largely of short-answer and essay prompts.
What makes this all worthy of discussion is that the usefulness of rote-memorized data in the real world has, by and large, come to its end. The uselessness of rote-regurgitation of grammar+vocab is not just a "thing" of language courses anymore. Far more prominent than that: it's a "thing" of programming courses.
Programming, as a profession, is defined by requiring exactly the sort of free-form thinking and expression as natural-language communication. Lots of people get out of university as perfect engineering students, with a large rote-memorized bag of algorithms and syntax under their belt. These people are the reason most industrial companies want to see prior coding experience of some sort before they will hire you as a software developer: their training sucks and their work sucks.
Research science actually works much the same way. You cannot regurgitate what you were taught in coursework and ever hope to succeed as a scientist. I'm quite personally acquainted with an adviser other than mine who has had to "Master out" some grad-students who did exceedingly well in their coursework but just couldn't do original research.
What real professions actually require rote-regurgitation anymore? Near as I can tell: investment banking... and investment banking, and only mostly because it's considered a way of hazing the newbies. In most other cases I've heard of, the rote portion of work has been automated to cut down labor costs, and what remains is either nothing, mostly physical work, or requires enough skill and independent thought to go beyond what a rote-trained person can do. Even the "most rote" of all current occupations, the "lowly" burger-flipper, actually needs a lot more coordination and communication ability than the engineering-school philosophy will give you. No, really: burger flippers have to be able to do many things without being prompted by a proctor. They have to be able to talk to customers in their own language.
That's why it seems such a paradox that a place responsible for so much excellent research and entrepreneurship teaches so completely according to this method.
The funny part is that after leaving undergrad, I had never imagined the thing I would dislike most about grad-school would be the coursework. It's not horrific and I'm certainly not incapable of it. It's just run entirely according to the Engineering School Mentality, which results in fairly bad teaching methods being used and a whole lot of antagonism between faculty and undergraduates, in which us grad students basically have to "pick a side, we're at war".
What is the Engineering School Mentality? "What matters is the grade, specifically on the final exam." It's a fairly different philosophy from what I saw in undergrad at UMass Amherst, and I'm somewhat disturbed by its... outgrowth into a full-blown teaching method. Its full expression is best captured in a veritable proverb I heard from one professor: "Students will do anything for a high grade on the exam; in extreme cases they'll even learn."
Let me explain. Technion does not handle coursework in the fashion of an "American-style" university, with 50-90 minute lectures several times per week per course and a steady stream of homework assignments, quizzes, in-class participation requirements, and mid-sized exams. Instead, it books one 3-hour lecture per course per week, gives a comparatively small amount of homework, requires only a moderate amount of in-class participation, and then puts everything riding on final exams and semester projects.
Exams period here is two months: one month to attempt your exams the first time, and the second month to attempt them again because you didn't do well enough. Of course, systems like this have "worked" for decades in engineering schools around the world. They certainly produce results: when "only the strong survive", you at least guarantee those who survive are probably pretty strong
What makes this a subject worth writing about at all is the bizarre contradiction of applying a teaching method designed for mechanical engineering to, of all things, a natural language class. As a foreign-born graduate student, I have to take and pass the course "Reduced Hebrew", which despite its name, is only given to those who already possess a moderate level of conversational spoken Hebrew. And it is ridiculous.
There seems to be little logical ordering to the curriculum, and the vocabulary and grammar taught are exclusively those of academia. For example, we learned words like "journal article" (ma'amar ktav-et) before we learned participles of common verbs. Worse, the teacher forbids herself from ever switching out of Hebrew to explain anything, resulting in our having to try to learn adult, academic Hebrew based on the pidgin dialect of small children and new immigrants. It also gives us continual utterances of "exactly the same thing" (bidyuk oto davar) or "X is like Y" (x zeh kmo y) for pairs of words or phrases that, when you try to actually use them outside class, mean something subtly but significantly different.
Worst of all, the course consists of zero free-form speech or writing. Question: what kind of language course enters its fourth month without one lengthy writing assignment or in-class unstructured conversation? Answer: a language course invented by an engineering school. The complete and total emphasis is on precisely memorizing grammatical forms and vocabulary sets for accurate regurgitation in exercises and exams. Any attempt to actually speak the language is not merely unrequested but almost punished: the most common response is, "We don't say it thus" (Anachnu lo omrim et zeh kakhah.). Colloquialisms are taught rarely, if ever, and when mentioned are discouraged with a disturbingly satisfied "It could be, but this is the prettier Hebrew" (Yekhol lihiyot, aval zeh ha'Ivrit yoter yafah.). The native idioms and metaphors of Hebrew, or matters of local culture, have not come up, ever. Speaking becomes more of an opportunity to "lose points" and interrupt the flow of the curriculum than to communicate one's thoughts.
The result is that I can feel my brain malfunctioning: my ability to read, hear, and understand Literary Hebrew (as opposed to the less Aramaic, more Arabic-and-English "street Hebrew" spoken colloquially by much of the population) has grown quite significantly in the time I've taken this course... while my ability to speak or write the language on my own has only scarcely improved.
And what's the point of such a course as "Reduced Hebrew"? It's what you might have guessed: to pass an exam that fulfills a university requirement. This seems like the textbook example of premature optimization, or optimization on the wrong criterion: "We have a metric, and we optimize for that metric. It's up to the
I mention all this to contrast it with the ulpan system of language learning, the normal mode of immigrant language-absorption in Israel, which chiefly emphasizes free-form conversation and writing. And it works! A diligent ulpan-goer with even the slightest ability for languages will usually significantly improve their conversation and writing in even the five months of government-provided absorption ulpan, speeding up the absorption process and enabling new immigrants to join society. There are no exams until the very end, and the final exams consist largely of short-answer and essay prompts.
What makes this all worthy of discussion is that the usefulness of rote-memorized data in the real world has, by and large, come to its end. The uselessness of rote-regurgitation of grammar+vocab is not just a "thing" of language courses anymore. Far more prominent than that: it's a "thing" of programming courses.
Programming, as a profession, is defined by requiring exactly the sort of free-form thinking and expression as natural-language communication. Lots of people get out of university as perfect engineering students, with a large rote-memorized bag of algorithms and syntax under their belt. These people are the reason most industrial companies want to see prior coding experience of some sort before they will hire you as a software developer: their training sucks and their work sucks.
Research science actually works much the same way. You cannot regurgitate what you were taught in coursework and ever hope to succeed as a scientist. I'm quite personally acquainted with an adviser other than mine who has had to "Master out" some grad-students who did exceedingly well in their coursework but just couldn't do original research.
What real professions actually require rote-regurgitation anymore? Near as I can tell: investment banking... and investment banking, and only mostly because it's considered a way of hazing the newbies. In most other cases I've heard of, the rote portion of work has been automated to cut down labor costs, and what remains is either nothing, mostly physical work, or requires enough skill and independent thought to go beyond what a rote-trained person can do. Even the "most rote" of all current occupations, the "lowly" burger-flipper, actually needs a lot more coordination and communication ability than the engineering-school philosophy will give you. No, really: burger flippers have to be able to do many things without being prompted by a proctor. They have to be able to talk to customers in their own language.
That's why it seems such a paradox that a place responsible for so much excellent research and entrepreneurship teaches so completely according to this method.
0 commentaires:
Enregistrer un commentaire