Blog Post

The Invention of Failure

What if we eliminated "flunk out courses" that were defined as "rigorous and demanding" and, instead, set our goal as ensuring the success of every student in equally "rigorous and demanding" courses?   That is, what if we decided what we considered to be the high bar of excellence in a given course and then not only agreed that we could give A's to anyone who achieved that goal but worked, collectively, to find the right tools, methods, and partners so that each student in the class could theoretically achieve an A--or at least a "pass" for the course?   Do you see how setting the goal as success rather than failure changes everything about success and learning in the course?   


I am not saying that every student would, in fact, pass.  But if the goal were to work with every student willing to work hard  to pass and to find the best way for them to pass, then the whole constellation of ideas based on "rigorous means failing a lot of students" would fly out the window.   "Grade inflation" (in all its unfair and hypocritical ramifications) would not exist.  The metric would be "grade attainment"--and the prof would be a success if the students were.  The whole learning dynamic changes.


Also, if one were not grading students on a curve, with a presumption that a certain percentage would or should fail, the one could begin by thoughtfully defining the terms of success .  Thinking this through at the beginning and staring the goals before the course begins, eliminates the mysterious part of learning for those who do not have the basics of learning how to learn.  Transparency changes the energy  of the class so students can work towards reaching a goal--not psyching out and guessing what the teacher may or may not want. 


I am currently working on the third chapter of my new book on the history and future of higher education, "The Invention of Failure."   I am fascinated by all I am learning about statistics, testing, IQ, multiple choice, grading, class and school ranking, and failure in the Industrial Age, and the clear relationship between learning failure and Taylorism and all the aspect of scientific labor management of the era.  

One of my arguments is that everything in education--kindergarten to lifelong learning--is essentially configured on an implicit curve with the high point of the curve implicitly defined by the Top 10 elite universities that serve approximately .03% of the current college population.   Accreditation, credentialing, and everything else uses that elite norm as the standard and everything is measured from that goal.  It makes elitism--making it through the very tiny eye of the academic needle--the true metric of higher education success.

I've not yet read Lani Guinier's new book on testing but I will be interested to see how much her work and mine intersect.  I suspect there will be some real confluences.  In the meantime, I want to work this semester, in our "Mapping the Futures of Higher Education" course, to set up the goal, for our graduate students and for their undergraduates, of a rigorous, difficult, exactly course--and the other goal of finding the right methods, tools, and partners so that everyone succeeds in attaining a passing grade in that exacting, difficult course.  

Success when the bar is set high gives you not just immediate satisfaction but a toolkit for success when you confront obstacles later--in school and out.



My syllabi are designed so that if students attend class regularly and do the homework, they are guaranteed a 70% (2.0) in the course. Of course, if students attend regularly and do the homework, it is likely that they will do much better than a 2.0. I do gave a lot of high grades, but they are earned grades.  My classes are demanding--but they are designed both for rigor and for success.


I am going to build this into our "Mapping the Futures of Higher Education" syllabus-building exercise next week, Steve.  Thank you.


I agree, Cathy. I think we have an obligation to validate the quality of the work, so I don't create grading systems that simply add up points, and I don't give "extra credit." I do, however, allow students to revise and resubmit based on feedback, and they earn the second grade. And I offer students the option of proposing other ways to demonstrate acheivement of the objectives (although I've never had someone actually take me up on that). I think those strategies speak to your suggestion that we should work with students to help them to acheive a high bar.


This afternoon, I had an interesting discussion with a student who is having difficulty with his self chosen subject because of terminology he is finding to be extremely difficult to understand.  He was considering changing his topic because he was afraid of failure.  In a more tradition class, his fears would be justified; but not in my class.

I assured him that I would prefer that he challenge himself with the topic in which he was interested instead of changing to a safer topic.  I also assured him that even if he could not master the terminology—that even if he had a spectacular failure—that he could still earn a 4.0 in the course.

I don’t have a clue about his topic so he cannot expect any guidance from me while he struggles with the technical terminology.  But the only thing he really needs from me is to remove the possibility of failure in the course

I have no doubt that the student will master the terminology.  But even if he doesn’t, he will learn more from “failing” on this project than being “successful” on a safer choice.

Out of respect for the student’s privacy, I cannot provide details.  However, I can state that he can met the course objectives even if he “fails” this assignment; that his 4.0 will be earned and not a gift.

On a final note:  After our conversation, I made the student read Dr. Davidson’s “The Invention of Failure.”