Is ‘more better’ a mistake if Shakespeare said it?

Research Associate Sean Murphy looks at Shakespeare’s use of more better, and considers whether we can really consider it a mistake if Shakespeare himself said it…

Learning English as a foreign language is hard. Learners are often corrected by teachers for saying or writing things like staying in our country is more better than going abroad. Being corrected by the teacher in front of their peers (with the corresponding loss of face which that may entail), or seeing the error underlined in red, can undermine learners’ self-confidence. But what if the teacher pointed out that, in Shakespeare’s First Folio, we find:

Art ignorant of what thou art. naught knowing

Of whence I am: nor that I am more better

Then Prospero

(The Tempest, I.ii)

The teacher might also highlight that, where the student uses the modern standard than, Shakespeare uses then. Of course, whether than and then after a comparative form were really different words, or merely orthographic variants (the First Folio contains 5 instances of better than, and 78 of better then) is debatable. Shakespeare uses more better three times in his plays, and in the very large database, Early English Books Online (comprising a wide variety of texts from 1473 – 1700), more better occurs no fewer than 5,180 times. In other words, rather than ‘making a mistake’, the student has used a structure common in Early Modern English (more better), together with the form (than) commonly used in comparative structures in present-day English (PDE).

As part of the Encyclopaedia of Shakespeare’s Language project, I’ve been re-reading the plays and regularising variant spellings in the First Folio (see the post on Smoothing out spelling variation). Having worked for many years as an English language teacher, I began to notice how many common ‘mistakes’ made by Spanish learners of English crop up in Shakespeare. Among others, these include non-standard use of: comparative and superlative forms of adjectives (more better, most coldest); prepositions (reason of, despite of); nouns (these news, informations) and double negatives (I can’t wear no dress)[1]. I amused myself with the idea that perhaps, as one writer has suggested, Shakespeare’s plays were actually written by Cervantes[2]. A more plausible explanation seemed to be that here was evidence of change in the English language, and the occurrence of a process of standardization from Shakespeare’s time to the present-day.

So to what extent can some typical learner errors be found in Shakespeare and Early Modern English? Using the power of corpus linguistics, and databases available on Lancaster University’s CQPweb (a corpus query tool)[3], I was able to search for examples of the above-mentioned non-standard forms in the following corpora: the Longman Learners Corpus (9 million words); the Shakespeare First Folio Corpus (1 million words); and the Early English Books Online (EEBO) Corpus (1.2 billion words). The results, all of which are attested examples in the corpora, are given in the table below.

Common non-standard forms in learner English compared to Shakespeare and Early Modern English

*   search for despite and despight

** search for news and newes

The first point to note is that all the learner ‘mistakes’ searched for (in bold) can be found in both Shakespeare’s plays and Early Modern English. The first statistic given means that there are 133 instances of more better, more stronger, more richer, etc. in the Longman Learners Corpus. When this figure is normalised, we get a frequency 14.8 instances per million words. As we can see, this is less than the normalised frequency in Shakespeare, but more than in Early English Books Online. In fact, if we glance over the normalised figures as a whole, we see that Shakespeare uses the non-standard forms (i.e. non-standard in PDE) more than the learners or his contemporaries, except in the case of reason of (where EEBO wins by a mile) and informations (the learners comfortably take the biscuit).

Secondly, it is important to remember that Shakespearean texts were written before any kind of written standard form of English had taken root. Standard forms simply did not matter so much in the late 16^th and early 17^th centuries as they do nowadays. With the advent of a standard form of the language and increasingly prescriptive attitudes from the 18^th century onwards, adherence to a standard is now considered to be of great importance, and non-standard forms are often regarded as sub-standard. The paradox with Shakespeare is that there are forms in Shakespeare that are non-standard by today’s conventions, yet are lauded as a kind of supra-standard.

With language learners and teachers in mind, we might reflect on the following:

• if some learner ‘mistakes’ can be found in canonical works such as those of Shakespeare and other writers, shouldn’t they more aptly be described as ‘variants’ – after all, they are the same as attested historical forms, some of which are regarded as prestigious;
• the negative connotations implied by the term ‘mistakes’, and its association with prescriptivist attitudes to language are out of sync with modern linguistic theories which aim to describe language use, whether in the press, by learners or by previous generations;
• English learners need to be praised and encouraged to increase their confidence in using the language – excessive attention to ‘mistakes’ is unlikely to help them achieve such self-belief:
• teachers can raise learners’ awareness of historical variation and encourage comparison with PDE, and may even spark an interest in Shakespeare.

In short, learners of English should take heart. Far from making mistakes, they are speaking like Shakespeare. Their English teachers should be proud of them.

[1] Of course, the double negative is still used in PDE, as this example shows: they wouldn’t have no records (BNC Sampler Corpus).

[2] The article is in Spanish. It reviews a book by a Catalan author who claims that Cervantes and Shakespeare were the same person, on the basis of an alphanumeric analysis of their works.

[3] Available at https://cqpweb.lancs.ac.uk/

[4] Statistics shown refer to (absolute frequency / normalised frequency per million words).