Having a Smooch with some Online Writing Tools

Whilst spending many hours in front of my computer writing, teaching and creating online courses, I must confess to never having given all that much thought to online writing tools and their pedagogical use. The systematic creation of a rubric in my blended or online courses entitled ‘Useful Links’, providing links to useful English language learning sites (see Resources Page for example) has been pretty much the extent of my advice to learners. Student usage of these tools, and any others they might come across themselves, could be taken as proof of learner autonomy and need not be an inherent part of course structure.

Listening, at the recent RANACLES conference, to Aurélie Bourdais (University Lyon 2) presenting her ongoing doctoral thesis on how French high school students use online translation tools, made me think it might be time to reflect further upon the issue.  Aurélie quoted Germain & Netten (2004) who claimed that systematically using such tools could hinder the process of language autonomisation (as opposed to learning or general autonomy) in school children learning French as a second language.

In April 2019 at the end of the blogging course I had asked 2^nd year students Paul-Valéry University students the following question (via a questionnaire): ‘What online tools did you use to write/check your English?’. They mentioned having recourse to Google translation, Linguée, Onlinecorrection, Reverso , Wordreference. To begin with, I decided to look at the two correctors mentioned rather than the online dictionary (Wordreference), or translation tools (Google and Linguée).

To check out the free correctors I fed into the correctors the following short extract from a text written by a French native speaker in the context of a creative writing exercise:

 ‘It was a day like every basic day in life of Livia. She was getting up, was showering, dressed and was leaving her home at 8am, to take subway and arrive at work at 8:45 am. Livia was secretary in estate agency in San Francisco since 1 year; when she moved from Chicago. She banal and paisible life.’

OnlineCorrection.com

This tool claims to check for spelling, and basic grammar and stylistic mistakes and enables the user to choose ‘dialects’ including American, British, Australian, New Zealand and South African English (in that order).

When I tested with the dialects British, New Zealand, South African English, 3 spelling errors were highlighted: ‘Livia’ and ‘paisible’. When I switched to American and Australian English only 1 spellling error showed up: ‘paisible’.

Reverso Speller

This tool claims to check grammar and spelling for English texts. You can choose between accepting spelling that is English, American or both.

When I fed the text in, Reverso automatically made 4 corrections, two of which were correct concerning grammar (missing article before ‘life’ and ‘subway’, one of which was incorrect concerning grammar (‘has been secretary’) and one incorrect concerning vocabulary (‘paisible’ which was changed to ‘possible’).

So, on the basis of this short text neither of these tools seemed to me particularly convincing. However, my curiosity was piqued, and I decided to see what else was out there. In his article ‘Second language writing online: An Update’ (2018) Robert Godwin-Jones refers to a number of other tools:

Grammarcheck

This tool picked up on one vocabulary error ‘paisible’ which it underlined in red.

Spellcheckplus

Picked up one vocabulary error ‘paisible’ and a grammar error on ‘1 year’

Grammark

Picked out zero errors but ‘wordiness’ was highlighted as being a potential problem.

Grammarly

To use this tool I had to set up an account but this was quick and easy to do. Before inserting my text there were 5 writing goals to set: Audience (general, knowledgeable, expert), Formality (informal, neutral, formal) Domain (free version = default setting normal, paying version = academic, business, technical, casual, creative) Tone (neutral, confident, joyful, optimistic, friendly, urgent, analytical, respectful) and Intent (inform, describe, convince, tell a story).  Tone and Intent are experimental goals not currently affecting feedback.  I set the following goals: audience = general, formality = neutral, tone = neutral and intent = tell a story.

The corrector came back with 6 alerts (three concerning the use of articles, one concerning the proposition ‘since’, one punctuation and one vocabulary ‘paisible’) all of which were valid. It was noted that the text was very clear and engaging and delivery just right. Two advanced alerts were given regarding punctuation and improper formatting but for more detail concerning these issues payment was necessary.

This was, I felt, quite impressive and I decided to investigate further.

I pasted in the rough version of this post’s first paragraph with the default settings (audience = knowledgeable, formality = neutral, domain = general) plus tone = confident and intent = inform. The corrector came back with 1 incorrect grammar alert. It considered my text to be  very engaging, very clear and its delivery just right (oof!). 5 additional issues (one of which apparently concerned ‘outdated language’ would have been available in the premium version. When I changed the audience (or the tone and intent) goals the same results were obtained. However, when I changed the formality setting to informal the text scored better and only 4 additional issues were flagged up.

Finally, I pasted in a short piece of creative writing (from one of my foreign language students) to which I had given full marks to in the past. The corrector’s feedback concerning engagement came back as ‘a bit bland’ a judgement which I absolutely did not agree with. How serious was this error of judgment in pedagogical terms? Had the student submitted himself his text to the online corrector would the negative feedback have adversely affected his creative confidence? My guess would have been not. As we become more adept as writers we have more confidence in our personal judgment. I’m pretty sure he would have done exactly as I did concerning my blog post: read the suggested correction, considered it and ignored it. And the chances of him doing just this would be increased if he had been trained in correct usage of this sort of tool.

Perhaps the domain goal ‘creative’, not currently available for the free version, will increase the tool’s sensitivity to more expert/original writing.

So, conclusions to all this? I think the only online writing tool I would actively recommend at this stage to my online students is Grammarly.

• The use of such a tool implies conscious reflexion on the process of rereading and rewriting which can improve writing.
• The tool picked up on simple grammatical errors which could be corrected before submitting work to peers or teachers for more feedback who can concentrate on more complex grammatical issues (eg. the problem with the use of the past continuous tense: ‘She was getting up, was showering, dressed and was leaving her home at 8am’).
• Even if the settings are neither fully operational nor that sophisticated,  it is clearly interesting in pedagogical terms  for students to be forced to reflect on why they are writing, for whom they are writing and how to modify their writing according to the effect they wish to create. Thinking about different genres and genre is an inherent part of all the writing courses that I teach and I would not suggest that online tools replace this course work.  But as we know teaching is often finding complementary ways of saying the same thing. Sometimes over and over again. If software can help this process then why not? Having to set the goals would remind students of these issues. And results could be analysed. Perhaps an activity could be proposed in which students submit their texts to Grammarly and simultaneously to their peers and then have a forum discussion in which they reflect upon the different feedback obtained. Alternatively, students (group work?) feed several texts from different sources into the tool and then discuss results.

To conclude, I don’t think that systematic usage is the main issue concerning the use of online writing tools by students. I think that what is important is finding ways of accompanying students on how best to use the tools, having them identify limitations and avoid the pitfalls of uncritical usage and incorporate valid corrections into the process of rewriting.

To be continued…the use of translation tools?

Other recommendations for pedagogical usage of online writing tools would be very welcome.

P.S. This semester, students added The Conjugator and Lexico to the list of tools that they used.

(April 2020 Update – students mentioned using translation tool deepL)

With thanks to fig_tart, instagram for the image.