Having been engaged in language study for over four decades, made attempts on four foreign languages and witnessed the growth of my first language, English, I can assure you that I have invested a great deal of time into learning vocabulary. Much of it was wasted. Much of it was spent learning useless words in ways that did not teach me how the words work in the target language. Vocabulary learning strategies were not taught. It was simply assumed that students would memorise context-free bilingual lists as if the L2 words worked in the same way as in L1. If I was lucky, my attempts to use the words in sentences and texts were returned bespattered with red ink highlighting collocation and colligation errors in particular. ​

You can probably think of some counter examples. And so can I. But what I see in contemporary course books does not negate most of the above. ​

We can task our students with identifying relationships between words and within words, and depict them meaningfully. 

These examples come from my Versatile Blank Book, which you can read about on this site.

Every layer of the hierarchy of language can be explored by students in a text. The exploration of Texts as Linguistic Objects (TALO) reveals how an author has used words and word forms, combined them as collocations and colligations, formed phrases and clauses that are linked with metadiscourse chunks to ultimately form texts. This is the bottom-up process that we employ both subconsciously and consciously when we speak and write. ​

As well as being linguistic objects, Texts are also Vehicles of Information (TAVI) which invoke top-down processes as we combine the content of the text with what we know about the world through schemata, general knowledge and our expectations of text types. In this way, readers and listeners are engaged in their own personal knowledge creation. 

Thirdly, Texts are Springboards for Production (TASP). We respond to texts by combining several texts on the same topic, by critiquing aspects of the text, and by discussing the potential impact of this new knowledge, for example.

To put texts under the microscope, VersaText is an open access, web-based resource that allows teachers and students to paste in a single text. The program provides several tools that foster discovery learning. The first tool is the word cloud, which depicts not only the relative frequencies of words in a text, but it colour-codes part of speech. The word cloud is highly customisable: the number of words, the choice of words vs. lemmas, which parts of speech to show. The relative sizes of words in the word cloud illustrate the extent of repetition in a text and repetition is the most commonly used resource to create lexical cohesion in text (Halliday & Hasan, 1976).​

The opportunity to put the language of a single text under such a microscope is invaluable to students of CLIL, EMI and ESP, as the texts are models of the language of subjects and fields that the students need to have a productive knowledge of, if they are to be acculturated into their subject disciplines. This is essential for TASP. 
​

Halliday, M.A.K. & Hasan, R. (1976) Cohesion in English.  Longman
Hoey, M. (1991) Patterns of Lexis in Text. OUP.
Johns, T., Davies, F. (1983) Text as a vehicle for information: the classroom use of written texts in teaching reading in a foreign language. Reading in a Foreign Language, 1 (1)
Partington, A. (1998) Patterns and Meanings: Using Corpora for English Language Research and Teaching.    John Benjamins.

In my last blog post, "…", I wrote:

• When we search a corpus for "does not a .* make", it is clear that this is a systematic linguistic exploitation, as the great British lexicographer, Patrick Hanks refers to such creative uses of language. 

Since I wrote this, this great British lexicographer has passed away.
 
Patrick's theory was articulated in his unputdownable book Lexical Analysis: Norms and Exploitations (2013). It is an empirical investigation into how words go together in collocation patterns and constructions to make meanings. In the book he writes extensively about patterns of normal usage and then demonstrates how a great deal of creative language use is also patterned, albeit differently. By the time Lexical Analysis was published, I had known Patrick for quite a while during our overlapping years at the Faculty of Computer Science (FI MU Brno). Our offices were on opposite sides of a corridor across which we'd often call to each other.

​Patrick had not been fully apprised of the doctoral defence procedure. Czech doctoral defences are preceded by an oral exam and he more or less delivered his inspired defence during the exam. We had to wait outside the exam room while the committee determined his exam result before allowing him to give his defence, which turned into a grand round table discussion. I remember James Pustejovsky, who'd been flown from the USA for the defence, ramping up Patrick's interrogation by claiming that he was getting off a little bit lightly. He passed.

Before the New Oxford, there was the COBUILD project at Birmingham University in the 1980s. Patrick was the managing editor of the COBUILD Dictionary. I distinctly remember the day he strode the metre and a half from his office to mine to tell me that he'd just learnt of the death of his dear friend and colleague, John Sinclair. Sinclair had run the highly innovative COBUILD project for a decade. Patrick would often relate the day Sinclair announced to his team of lexicographers that all of the dictionary's definitions would be written in full sentences unlike the standard practice of providing synonyms and snippets to express a word's meaning. These sentence definitions are built up of chunks of information that explain what and why and how something means what it means. This aspect of the COBUILD dictionary is why I recommend it to students and teachers in all my training and writing. I shall wax lyrical on this in another post.
 
One Friday afternoon, Patrick drove around the corner from the faculty to the bank, unnecessarily and lamentably. He came back from the bank on foot, his car having been stolen. Unfortunately, he hadn't finished unloading boxes of rare and specialist books from his car into the flat that a friend of mine let to him. The police came and drove Patrick and me around the neighbourhood for quite some time before they took us to the police station where we spent hours answering questions for the policeman to fill in forms. I was interpreting for him and I had to keep asking if cars really have all of these features he was being asked to say yea or nay to. It had been well over 20 years since I had owned a car and half of the features sounded indulgent and verging on sci-fi. He never got his car back let alone the books. But we did spend that Friday evening washing down a hefty steak with a fine Czech brew or two.
 
While still in Brno Patrick started a project that he pursued to the end of his working life. It is the Pattern Dictionary of English Verbs (PDEV), publicly available as an online database. He and his team studied concordances of verbs, tagging their grammatical and lexical features. They used Sketch Engine of course, since it was born and thrives to this day in the faculty where we were working. Sketch Engine achieved an international reach when the philosopher-linguist, Adam Kilgarriff, bought into it through a university–enterprise agreement. Adam died of cancer at the age of 55. In fact, the last time I saw Patrick was at Adam's funeral in Brighton.
 
Here is the entry for adapt in PDEV.

Every meaning/use of a verb has a unique pattern, which is why we know which meaning of a verb is meant when we read or hear it. And we use this information when we write and speak. Native speakers do this subconsciously while non-native speakers learn this, although to the best of my knowledge, there are no vocabulary teaching resources that employ this user-friendly description of meaning. I'm working on it! Watch this space. While an entry does not employ terminology such as valency, complementation or transitivity, it does use semantic types which are indicated in square brackets. The PDEV entry for answer has 14 entries, two of which are tagged phrasal: answer back and answer for and two are tagged idiom: have a lot to answer for and answer call of nature.
 
I participated in the PDEV corpus work for a short time. I was convinced that the searches could be partially automated as I had developed database scripts for automating Sketch Engine CQL searches for verbs in their known grammar patterns as part of my MA dissertation research. But Patrick preferred to work from raw data without skewing observations towards predetermined patterns. He also rejected intuition. There were times when I pointed out that the data did not show something about a verb that should be part of its description.
 
While his interest was in verbs, mine lay in nouns as the primary carriers of meaning in message clauses. I posited applying the PDEV approach to nouns, but Patrick had a completely different methodology for this part of speech. The cognitive profile, as he called it, consists of a set of short, objective sentences each sentence revolving around a collocation of the noun. For example, his spider profile includes these sentences:

• Some species of spiders hunt prey.
• Spiders lurk in the centres of their webs.
• Spiders have eight legs.
• Many people have a dread of or are frightened of spiders.

 
I have adapted this for teaching purposes, preferring the more student-friendly term, word profile. In pedagogical circles, a cognitive profile is more likely to pertain to a student than a word. I'll elaborate in another post.
 
I took him a couple of times to visit the old folks in Mikulov, a beautiful Moravian town on the Austrian border. The old folks were my dear friends, a Czech man and his South African wife, who had lived in my town, Sydney, for 15 years before moving to Czechia after the fall of the wall. We used to chat at the bar in their house, drink carafes of Moravian wine, play Noel and Cole on June's Steinway and stumble home the following day.
 
Patrick did invite me to stay with him for a few days in Bristol during which time he was frantically busy, but we did manage to talk about pattern descriptions of nouns, about his loneliness, his depressive moods, his ex-wife and his beloved daughter.
 
I can't imagine where my thinking about language would be had I not spent this time with Patrick. They were pivotal years for me, and it is a great honour to have been in his presence. You are very well-remembered, dear friend.