Error Correction: Why does it seem natural when we know it is a waste of time for acquisition?

Lately, I’ve discovered myself in moments where I correct students’ Latin errors in class.  You might already be thinking, “Correcting a students’ language mistake in a language class makes sense.”  But it never feels right, yet it seems to happen automatically.  But what purpose does error correction serve?  My assumption is that we all have experienced someone who corrects our native language use in this age of social media.  Honestly, teachers are the worst about this, especially when an administrator sends an email.  But why we do we go there?  What purpose does it serve when we correct someone else’s grammar?  In class, I seem to have this feeling that I am somehow moving students forward with their command of language when I start to correct an error.  But the way students react in these moments makes it clear that error correction turns on the affective filter.

But what do others say?  The topic has popped up quite a bit on different Latin teacher groups as of late.  The common argument in support of error correction by Latinists seems to be that we as instructors must put correct language in front of students.  The debate even moved toward self published novellas perceived to not be at the standard of correctness expected by instructors.  I realize this is a different issue, but I found it interesting that comments about error correction seemed to lead to an attack on modern Latin writers.  I agree that as instructors we need to use correct input for students but I’m left with 3 questions around error correction and input.

• Does error correction guide students uptake of correct input?
• Should I consider student output as input for other students whether it is correct or not?
• Are there any other variables that I should consider around the topic of error correction that I might not be considering?

The first question seems the most easy to answer.  There has to be research about error correction and student uptake.  Thankfully, colleagues in the Latin teacher group posted research to consider.  The first study I’m investigating is by Sheppard (1992).  1 ESOL group engaged in intense error correction over 10 weeks and the 2nd ESOL group received only content-oriented comments.  The 2nd group blew the 1st away with large effect size gains in comparison.  Extra input helped students become more accurate writers.

In another study by Bichener, Young and Cameron (2005) there were 3 groups of adult ESOL students.  The 1st group had explicit correction, student-teacher conferences and 20 hours of instruction.  The 3rd group had no error correction and no student-teacher conferences but only 4 hours of instruction.  Even with 16 hours less instruction, the no correction group produced slightly more accurate writing.  This study seems to point out that error correction may even be harmful to acquisition.

More studies demonstrate that groups who received error correction perform slightly below or relatively even with groups that do not receive error correction.  It seems a logical conclusion in such studies that, at a minimum, error correction is a waste of time.  Replacing the time spent with error correction with an extension of comprehensible input for the learner will be more valuable as it will aide acquisition.  Therefore, it seems an important conclusion that error correction is not worth the time and effort it takes to complete.  An instructor or parent is better off with a focus on new comprehensible and compelling input to offer new meaning for the learner to process.

Next, let’s explore question 3.  “What else should we consider in respect to error correction?”  Multiple studies (Kepner:  1991, Semke:  1980, 1984, Sheppard:  1992) demonstrate that corrected students tend to shorten and simplify their writing.  These students also avoid situations that they perceive might bring about error correction.  This is an important variable to consider within studies.  Even if corrected students perform somewhat similar to non corrected students, if they simplify and shorten writing one might argue that their proficiency actually decreases.  If this is the case, error correction literally stalls acquisition.  Once again, it seems that error correction is a waste of time.

The data above comes from “The effort of error correction on learners’ ability to write accurately” by John Truscott in the Journal of Second Language Writing 16 in 2007.  Truscott reminds the second language instructor to carefully analyze studies that center around error correction.  Other studies such as Russell and Spada (2006) may make claims that error correction produces better results.  But, more than likely, those studies derive results from students’ performance on grammar tests instead of performance on everyday use of language in writing or speaking.  Error correction might help students perform better on grammar tests, but grammar tests are not the final goal we hope to prepare students.  We hope they can communicate through understanding, speaking and writing the language, therefore it is important that studies test what we want students to do in the long run.

A 2nd article posted in a Latin teacher group led me to consider the effect of error correction on students’ affective filter.  Spanish Mama offers a post with suggestions for how teachers might increase their own proficiency of Spanish.  She interestingly starts this post with the following:

The insecurity mainly comes from people knowing I teach it. Talking with a bunch of old friends, who are native speakers? Fine. Teaching a low-level Spanish class? No problem.

But sit me down with another teacher who speaks better Spanish, and my proficiency literally crumbles in front of me. My heart races, and I can hear the mistakes falling out of my mouth. I want clarify: I do at least, hear the errors. Somehow I just can’t stop them!

Notice she points out that her anxiety rises around another teacher whom she perceives to speak better Spanish than her.  Our students play out this same situation in our classrooms everyday.  My conclusion is that students are already conscious that they will make errors as they enter my classroom.  Their perception of me as the expert teacher causes this before I even facilitate activities with Latin input.  Imagine how that perception will sky rocket when I focus on form and the correction of their errors!  No wonder students do not feel comfortable speaking or writing a language.  We stack the system in a way that feeds the power of the affective filter.

Spanish Mama continues with a description of how she feels about writing in the target language.

Another confession: I hate writing in Spanish on social media teacher forums. With friends I can text and message all day. But writing in a FB group, where the grammarians are just waiting to help me improve my Spanish? No thank you.

Often in my class I offer students an option of writing a response in a canvas discussion when I assess the interpersonal standard and expect a response from them in the moment of a class discussion or story.  Canvas isn’t the same as a public Facebook group but I wonder if students have a similar feeling as she describes above?  I’ve watched similar situations with Latin teachers who share a resource or a published Latin novella.  Colleagues continually have more to say about what is wrong with another teacher’s writing opposed to focusing on what is beneficial.  When the research is so clear, why do we default to error correction?  It doesn’t work and is simply a waste of time when considering how to help someone move forward in acquisition.

Spanish Mama continues:

I end(ed) up learning Spanish, teaching Spanish, and raising bilingual kids without planning any of it. My credentials are not all up-to-speed. I should be wide open to it, grateful to people who help me get better. 

My husband, a native Spanish speaker is like that. He insists on his English being corrected. Thanks to his thick skin, he has that strange ability to separate an honest mistake from his personhood. I get defensive when my story gets interrupted to let me know it’s tenga, not tiene; he’s grateful for the a-ha moment.

I’m going to disagree with her at this point.  Thick skin shouldn’t be required for us to be reflective about our language use.  This anecdote of the necessity of thick skin is another example of why error correction simply doesn’t work.  Think of all of the energy, motivation and time that is wasted in the process when someone trains themselves to have thick skin?  How much more progress would happen if instead that time and energy was focused on more input to continue the progression of appropriate language processing? Spanish Mama should be defensive, but not because of her potential errors.  Error correction based on form is a waste of time and we should stop doing it in all facets of language interactions.

My initial reaction to my second question is no, I shouldn’t consider student output to be input for other students.  This will deserve another post for investigation but my initial reaction is that I should be using student output to create more output from myself or other sources to continue the processing of input for students.  In this way, it doesn’t matter if students output is correct or not because the output from me will be correct that furthers the discussion or story.  But this new input may be more compelling as it is based on the thoughts and expressions of a student in the classroom.

As I come to the end of my thoughts I move from teacher to parent.  We have an interesting situation in our house.  Our kindergarten daughter, who is the middle child, is in a fairly new Spanish immersion program at her elementary school.  The program was started after our oldest, who is an 8 year old 3rd grader, had already progressed through kindergarten and 1st grade.  But she had the Spanish immersion kindergarten teacher before the program had been started.  This has created some competition but also it has compelled my oldest child to join Spanish Club when it fits into our schedule.  She is compelled by the competition she feels from her younger sister.

The affective filter is clear when our oldest child attempts to speak some Spanish around our middle child.  The middle child immediately corrects the oldest child.  Most of the time it is clear she is correcting and not listening to the message of her older sister.  As you can imagine, conflict arises at these points.  But the kindergarten student already seems to naturally want to correct the person she is speaking to.  I don’t think this is happening often in her classroom from the teacher but here it is at our dinner table, frequently.  And this is interesting because it is one area that our middle child can be the expert in that our oldest child cannot be the expert.  Is the error correction about Spanish or about a chance to be the expert?

Based on the little bit of research I’ve explored and the anecdotes to accompany the research, I’m convinced that error correction is not actually about helping people become better at language.  It is about something else.  Perhaps error correction is about multiple things.  Now that I’ve come to this point, what can I do in the classroom or with my own children to train myself not to use error correction?

• Focus on input that offers opportunities to build messages from the beginning of the input.  I have been teaching with a goal of comprehension for 5 years yet I still find myself falling back to old habits like error correction based on form.  It never hurts to continually remind myself that the message is the most important thing I do each day, no matter the situation.
• Take a deep breath when a moment happens and my instinct is to correct an error before speaking.  Is the situation an error for comprehending meaning or an error focused on form?  Hopefully this will allow me the time to process the situation before moving into error correction mode.
• As often as possible, keep track of moments that I move to correct an error.  When I find a pattern I will attempt to find a follow up Latin question that focuses on meaning instead of form.
• With writing assignments, ask Latin questions when the writing is not scored.  If it is scored for a grade, limit comments unless a student follows up as the grade seems to get in the way of the students’ reflection on the feedback in many cases.
• Work to bring myself to a position that I consider errors as moments in acquisition.  A student who says “Salvete, mihi amici” when she enters the room conveys her message even if “Salvete, mei amici” might be more likely to be said by a Roman.  A student who has recently acquired mihi with constructions like mihi placet or licet mihi will attempt to transfer the use of mihi to other situations.  Instead of correcting that student, which will move them to simplify and avoid the use of mihi, I should find moments to use the adjective meus, mea, meum more often to give them opportunities to acquire those forms as well.
• Attempt to combat the defensive mindset I find myself in when I see comments or situations in which Latin teachers correct each other’s Latinitas.  Instead, attempt to reframe those conversations to ideas around meaning and the understanding of meaning.
• Redirect my childrens’ conversation around Spanish to the meaning of what they are saying.  I will attempt to become the learner with them with the hope that they will focus on meaning to help me learn from them.

Awareness is key.  Whether it be in your classroom, with your own children, or especially on social media.  The next time you feel the desire to correct someone’s grammar, writing style or language use, just don’t do it.  No, really, don’t.  It’s a waste of time in respect to language use to focus on the correction of errors based on form. Reflect on the reasons you have this desire and refocus on what the person you are interacting with is trying to say.  The world will become a better place one refusal to correct another’s language use at a time.