• warm and fuzzy
• furry and fuzzy
• cuddly and fuzzy
• cute and fuzzy.

None of these eight combinations occur frequently enough in SkELL's billion word corpus to be anointed chunks. English can certainly boast many very high frequency “and” chunks, such as:

• first and foremost
• tried and tested
• doom and gloom
• chalk and talk
• part and parcel
• pen and paper
• trials and tribulations.

With their poetic alliteration and their internal rhyming scheme, these are chunks with their own semantic and pragmatic functions in the language.

Our fuzzy examples, however, are properties of the word. Make a word sketch for pretty much any noun, verb, adjective, or adverb in the English language and you will find words that and links it with. The more arcane the word, the more arcane the partners. See arcane, for example.

Search for words that are in a text you are studying or on a word list you are working with. Explore the "and" relationship by clicking on a word to see it in sentence examples. For example, in the "and" relationship of class, we see among others:

• gender
• category
• method.

Clicking on gender shows 40 example sentences that include class and gender. This must be more than a lexical relationship – it is cultural and political.

Clicking on category shows 40 sentences related to categorisation. Once again, this represents the way this topic is written and spoken about.

Remember that we do not need to read and grasp the sentence examples that corpus searches yield. A concordance page is not a text – it is a source of language data that we skim and scan to identify patterns of normal usage from which we learn how the language works so that our use of the foreign language approximates that of the many native speakers whose output has turned up in the corpus.

The "and" pattern, connecting two words of the same part of speech and with similar semantics, often offers lexical support. Rather than expressing something new, this quasi repetition reinforces the meaning or general impression being communicated. This is particularly noticeable with subjective and emotive words. Look at the "and" relationships of:

• angry
• funny
• modest

There is also the matter of word order. Words in these and relationships are usually said in a set order.

• Coffee and tea is much less frequent than tea and coffee.
• Compare and contrast vs. contrast and compare.
• When and where vs. where and when.
• Reading and writing  vs. writing and reading.

As language teachers and speakers of foreign languages, we are always interested in the properties of words. Knowing a word’s patterns of normal usage is key to using words as native speakers do. And the confidence that emerges from this knowledge impacts on fluency.

How often have you seen claims like these?

• Speak like a native in no time!
• Effortless fluency guaranteed!
• Learn a language in your spare time with no effort!
• Language fluency made simple and quick!
• Master any language with this one simple trick!
• Learning a new language is a breeze!
• Fluency in just 10 days!
• Master a language while you sleep!

Who would make such claims? Certainly not someone who has achieved a high level of competence in a foreign language. As an eternal language student, and a language teacher and a teacher of teachers, I am pretty certain that you and I have worked hard to get to where we are in our foreign languages. It was not effortless, it wasn’t a breeze and no progress was made while we were asleep. I was so desperate to improve my vocabulary, that I used to sleep with my dictionary under my pillow. No I didn’t, but that is the impression you get from some of these slogans.

Learning a language is hard work and there ain’t nothin’ wrong with hard work. Never confuse hard work with hard labour. Hard doesn’t mean boring and monotonous and it doesn’t mean frustrating and unrewarding.

The hard work we do when learning a language involves planning, monitoring and revising. It involves understanding what we are learning, which in turn involves connecting what we already know with what is new to us. Hard work involves practising what we have learnt so that it becomes automatic. Hard work involves using our time efficiently, choosing approaches that work for us. This requires us to assess or critique the approaches to language study that are introduced to us if we are fortunate enough to have various approaches. It is worth reflecting on how many different learning experiences we are having while studying – the learning #affordances of an activity.
​
For example, if we are learning a set of English words with their first language equivalents, whether in written lists, on paper or electronic flashcards, in a computer game or being tested by our study buddy, the only connection we make is between the L1 and L2 word. We do not learn how the word is used. This leads us to assume that the L2 word is used in the same way as it used in L1 and this is okay when it works. It is not okay the rest of the time. Psycholinguists refer to this assumption as the semantic equivalence hypothesis (Ijaz 1986, Ringbom 2007).

Another issue with learning L1–L2 pairs is the mental processing of the L2 word: what is your mind doing whilst trying to remember a word? Lower order thinking does not make for a rich learning experience. When we are critiquing our approaches to vocabulary study, we need to consider how many different features of words we are learning at the same time.
​
However, when we study the vocabulary of a text in a text, we see how keywords are used differently each time they are used. Yes, their different uses create different messages which means that the author is telling us something new about the keyword each time it is used. These different messages involve different words, which means we can make a diagram of a keyword as it is used in a text. I call these diagrams Word Constellations.

We can identify a key word and highlight all of its occurrences in the text, then highlight the words that are used with it. I prefer to do this with #VersaText because it is easy to see the left and right cotexts of keywords in a concordance. You can do this with at least several key words in the text. ​

In critiquing this approach, we need to consider how much the students learnt during the time spent.

• Did the knowledge gained empower them to use the keyword and its cotextual words better than learning such combinations from flashcards or bilingual lists?
• Did they meet any combinations of words that were new and useful?
• Did they have to check their dictionary for anything?
• Did they notice anything about the distribution of the keyword through the text?
• Did the process heighten their understanding of the text?
• Did drawing the word constellations feel like a strong learning experience?
• Was it an enjoyable process?
• Can they imagine their vocabulary notebooks full of word constellations?
• Are they likely to refer to them again? And again?

If they didn't create a word constellation of a key word in a text, what did they do? How did they spend their time? What learning took place? 

Once we have a keyword in its multiple cotexts, we can use them as the bases of our own sentences. We might like to use them to form questions to discuss with our study buddy or our favourite AI tool. Here are some simple examples of a chat with Perplexity.ai.

Hi. I'm a B1 student of English. Will you be my study buddy today?
Of course! I'd be happy to help you with your English studies.

• Are Czech walking trails a complex system?
• Are they connected together?
• Do they run through the whole country?

Is this time-consuming? Is it a good use of our time? Are we having strong learning experiences? Are connections forming in our minds that have a high chance of becoming permanent?

I used bilingual lists for many years, long after I needed to. I think that if I were to start another foreign language, I would need them as a beginner. But applying what I have actually known for a long time, I would move as quickly as possible to studying words with their natural cotexts.

It may be the case that people who make claims like those at the top of this article do actually teach vocabulary in cotext, but from the courses and resources that I have seen over the years, this does not seem very likely.

If you or your students ever create word constellations, I’d love to see them. And I would love to know what the process led to.

Feel free to join the VersaText Facebook Group where you can share your experiences and learn from others.
​

Ijaz, I. H. (1986). Linguistic and cognitive determinants of lexical acquisition in a second language. Language Learning, 36(4): 401-451

Ringbom, H. (2007). Cross-linguistic similarity in foreign language learning (Vol. 21). Multilingual Matters.
​
Veselá, Z. (2003) Czech Republic’s unrivalled system of marked walking trails. Radio Prague International, 5.12.2003.

I had an email from a teacher who loves using my VersaText tool with his students. In addition to the very welcome and rarely received praise for VersaText, he was enquiring into the possibility of adding a collocation feature. As you know, VersaText works with single texts, its slogan being, “learning language from language, one text at a time”, collocations are vanishingly rare. In fact, “vanishingly rare” is a strong collocation in English. Check out the examples in #SkELL.

#Collocation is defined variously. First and foremost, collocation consists of two content words of different parts of speech. Compound nouns and adjectives, phrasal and delexical verbs are not collocations. And neither are words that combine with that/ -ing / inf / wh-/prepostions. These are colligations and offer very little choice, if any. You’ve all seen gap fills in coursebooks and exams that test this. Collocation does permit some variation, but within limits of acceptability if you are going to use the patterns of normal usage of the language.

One category of definitions of collocation revolves around statistical frequency. These definitions rely on the number of times words occur in close proximity to each other. The verb collocates of trouble, for example, occur frequently within four words before and/or after the noun in SkELL's huge sample of English.

Other definitions of collocation are phraseological: cause trouble is the core of a clause, which is the essential structure that creates Messages, which in turn constitutes text. Up the Hierarchy of Language we go! 

Most key words in most texts collocate with different items because the author is telling us something new about the word. And this is why a collocation tool in VersaText would be by and large redundant.
​
One thing we can be sure of in a text is that the author is not going to repeat the same message repeatedly, again and again, over and over, unless they have some rhetorical reason for doing so.  Here is an example. In VersaText’s sample text, Learning Zone (a transcript of a TED Talk), we see that the verb spend is frequently used with time, and with other time words, e.g. minutes, hours, our lives. It occurs 13 times in the text. Time occurs 28 times in the text and is used thus: ​

Go to VersaText, select the Learning Zone text from the list, then click Wordcloud at the top. If you want the lemma of improve, for example, choose the lemma radio button under the word cloud. Click on any word to see its concordance in this text. This motivates many discovery learning tasks for the students.

If you want to learn more about studying and teaching English with VersaText, click the Course button at the top of the VersaText pages.

My phraseological approach to collocations in single texts is the Word Constellation. See my blog post linked below.