The Loom of Verbal Reasoning

“Immersion in the particular proved essential for the catching of anything general.” Albert Hirschman

Who’d have thought that empirical research on the Loom of Form and Meaning would be possible?

The purchase receipt is still in the book. I bought Essentials of Verbal Reasoning by O.B. Gregory in July 2010 from Foyles Bookshop in London’s Charing Cross Road, on impulse, during a lunchtime browse in the joyfully semi-random way that only other browsers in Foyles at the time may understand. Perhaps it’s different there now, for better or worse.

First published in 1963, the 2004 edition I have has no sales puff on its unattractively abstract cover. I remember browsing through and thinking there were some nice analogy problems in it, and especially seeing for the first time ‘hand : palm :: foot : X.’ But back then, once home, it sat on my bookshelf like an unexploded firework, for 8 years.

But when I was recently working on the Loom of Form and Meaning to categorise different types of comparison according to whether they involve forms and meanings which are the same, similar or different, I landed on this book as a source of raw data to test out my framework. This was to be the anvil upon which I would hammer my speculations and forge my Loom. Since then, this slim volume has been a testing ground which has fittened up my ideas about the Loom no end.

The book has 2,000 verbal reasoning questions in total, grouped into 200 exercises of 10 similar types of question, and further loosely grouped into similar exercises. Exactly how similar these questions, exercises and groupings really are, and what kinds of reasoning are represented, is the subject of my interest. It’s fair to say that not all verbal reasoning questions fit comfortably onto the Loom, but by far the vast majority do, and overall I found most types of connection represented in the exercises.

So what kinds of exercises typify each category in the Loom?

A1 (same form, same meaning) is represented by exercises involving definitions and characteristics (ex. 37 and 38, e.g. 37.9: choose ONE word a table always has: legs, cloth, flowers, chair), and comprehension puzzles (ex. 137 and 138, 165 to 170) where hidden or disguised characteristics of a situation must be identified by application of logical reasoning.

A2 (similar form, same meaning) is the most common type of exercise in the book, including finding the next word in an ordered sequence, sometimes involving meaning (ex.1 and 2, e.g. 1.3: choose the next word after trio, quartet, quintet… from sextet, band or orchestra) and other times only syntax (ex. 1.5: choose the next word after ask, bun, cry, dog… from cake, eat, puppy, weep), reordering a list (ex. 31 to 34, e.g. 32.1 to identify which word is out of order: letter, sentence, paragraph, chapter, word), restoring coherence to a paragraph by adding words from a list (ex. 43 to 46) or rearranging sentences (ex. 161 to 164), identifying the same letters included in different words (ex. 73 and 74, e.g. 74.1 to find two letters included in each of secret, plaster, storage, terrace, turnip), restoring the correct meaning of a word by adding missing letters (ex. 97.7: put a letter for each dot to make a complete word meaning goods carried by a ship or train: f . . . . . t), identifying the frequency of letters in words (ex. 139 to 142, e.g. 141.9 to find a letter which occurs twice in artificial and once in recitation), rearranging a given list of words in alphabetical order (ex. 145 to 154), using the letters in a word to match a given definition (ex. 155 to 158), letter rearrangement puzzles where meaning is irrelevant (ex. 159.2: which ONE word is not formed using the letters of the word SURRENDER: need, send, deer, reed, read), restoring coherence to a sentence by rearranging words (ex. 187.1: the repairs gallery was closed for art), adding or removing a word or part of a word to leave a coherent sentence (ex. 189.6: the flags were waving in the still breeze). All involve modifying but not completely changing a form to restore or continue meaning, coherence or order to the whole.

A3 (different form, same meaning) exercises play with synonyms (ex. 19 and 20, 35 and 36, 47 and 48, 108 to 113, 120 to 123. e.g. 19.3 to choose one word from {dry, hot wet} and {damp, water, cold} with closely similar meaning), reexpression with codes (ex. 91.10: if BFDBWIP means CRICKET, what does PDBWIP mean?), conversion between indirect and direct speech (ex. 173 to 176), and use of pronouns and more precise vocabulary to reexpress a phrase (ex. 179 and 180).

B1 (same form, similar meaning) exercises play on well established ambiguity and double meaning, where the same word has more than one related meaning through a shared etymological origin (ex. 75 to 80, 127 and 128, e.g. 77.2, where the answer SCALE means a series of musical notes, and a small plate on the outside of a fish, with both meanings being related to divisions, and 77.3, where BAND means a strip of cloth encircling something, and a group of musicians, with both meanings being related to a force that unites), and reuse of words in compounds where the main sense of the original words is preserved (ex. 133 to 136, e.g. 133.4 where joining cow and slip makes cowslip, etymologically meaning ‘cow’s slobber’). Gregory’s exercises here are limited to finding only the homonyms, rather than going on to explore the underlying unifying concepts between the two original ideas, which would be more of a B3 type exercise.

B2 (similar form, similar meaning) exercises are well represented, and involve ‘family relationships’ in the sense of a set of related words, though not in any particular order or structured relationship, including restoring coherence to a group by selecting elements unlike the others (ex. 3 to 18, e.g 5.8: dragon, phoenix, tiger, griffin, unicorn) or finding related words among others (ex. 23 to 26, 105 to 107, e.g. 23.1: which two of these words are related and different from the others – cat, kangaroo, bear, dog, giraffe), adding a similar word to a group (ex. 55 to 58, e.g. 55.1 to add a similar word to major, lieutenant, captain, colonel), and using the correct part of speech of a word to complete a sentence (ex. 181 to 184, e.g. 183.7 to complete with a noun derived from ‘receive’: the cashier gave me a ……..).

B3 (different form, similar meaning) exercises play on structured relationships such as opposites (ex. 21 and 22, 27 and 28, e.g. 27.9 to identify the word with opposite meaning to ‘minute’ from second, huge, hour, big), and various kinds of analogy (ex. 41 and 42, 49 to 54, 59 and 60, 81 to 86, 101 to 104, e.g. 41.9 to choose a pair of words with the same relationship as ‘tractor is to farmer’ from: sauce, saucepan, food, kitchen, cook).

A few of the B1 exercises involving homonyms include questions which should properly be classified as C1 (same form, different meaning), where there is no known semantic similarity or shared etymology (e.g. 77.4, where the answer RACE, meaning both a competitive test of speed, and a tribe or nation, is an etymologically unrelated homonym). Other than these, I didn’t identify any exercises in the C1 category involving more creative and imaginative interpretation of a given form, which falls outside the book’s stated scope of verbal reasoning.

C2 (similar form, different meaning) is fairly well represented though, for example with rhyme anchoring one aspect of a form while meaning is allowed to range widely (ex. 29 and 30, 69 to 72, e.g. 29.9: identify which of the following rhymes with reign… wren, ran, plain, brine), making a new word by taking out a letter (ex. 93 to 96, e.g. 93.9 to make a new word by taking one letter from butter), finding an anagram with a given meaning (ex. 143 and 144, e.g. 144.10 to rearrange the letters in ‘stop’ to get an upright piece of wood), and completing a sentence with any suitable phrase (ex. 199 and 200, e.g. 199.4 complete the sentence: We thought ….. but we found …..).

Gregory’s examples don’t include heterographs (e.g. what sounds like a word for animal flesh, and means to touch or join), or heteronyms (e.g. what written word is a heavy metal, and can be pronounced differently to mean ‘to go in front’).

C3 (different form, different meaning) exercises involving creative and contrasting juxtapositions, disjunctions, oxymorons and contradictions are absent, and like some types of C1 comparisons are outside the scope of the book. However, if I’d classified some of the above B2 exercises involving family relationships as their flipside, ‘odd one out’ exercises, I would have had some examples.

Here are some of the assumptions I made in my classification.

• If meaning is irrelevant to the solution, and it could be solved by a non-English speaker who only knows the order of the standard English 26 character alphabet, I’ve classified the exercise in the first row of the Loom where meaning is equivalent, because for the purposes of that solution the meaning doesn’t matter (ex. 53.3: finding the next word after tip, pit, part, trap, reed… doesn’t need any understanding of English to be answered correctly).
• If the exercise is solved by adding more individual elements (ex 1.1 to continue the sequence one, two, three, four… with five) I’ve classified it as A2 because the solution has added a new element to the original form of the sequence without changing its meaning, but where a general rule is required as a solution I’ve classified as A3, because it uses a totally different general expression of the same series rather than an incremental addition (‘the sequence of positive integers’ for the same ex. 1.1).
• Exercises can be considered from complementary perspectives which affects their classification, for example correcting or removing exceptions from a sentence or list can be seen as either changing an existing (B) or restoring an intended (A) meaning, or changing (2) or restoring (1) form. In such cases, I’ve classified exercises according to their end state, for example exercises where an intended meaning is restored by modifying a form are classed as A2.

Some groups of exercises are heterogenous:

• Ex. 87 to 90 involve individual anagrams (A2) but taken as a whole each group of 10 has solutions from a particular family such as vegetables or birds (B2).
• Ex. 124 to 126 are essentially B2-type rearragements of a list by pairs, though the nature of the individual pairings ranges includes opposites (B3), synonyms (A3), and family relationships (B2).
• Ex. 129 to 132 compares two different words (so column 3 on the loom) though the comparison may be row A if they’re found to have the same meaning, B if opposite (i.e. a structured relationship), or C if entirely unrelated.
• Ex. 171 and 172 require reordering of words into a coherent question (A2) and then answering the question directly (A3).
• There are also occasional exceptional individual questions within some exercises.

The full classification of exercises on the Loom of Form and Meaning

Hopefully it’s understood that my analysis is predetermined by the conceptual framework I’m testing. Isn’t it always the case? But what have we learned about the Loom anyway, and about verbal reasoning as presented in Gregory’s collection?

Verbal reasoning can be looked at as an exercise in comparison of shared elements of forms and meanings. From this perspective, the Loom has proved adequate as a way of classifying verbal reasoning exercises, though some parts of the Loom involving free interpretation (C1) and creative contrasts (C3) largely fall outside the ambit of a book about verbal reasoning. That said, question types intended to have a unique answer can be answered more creatively and amusingly (ex. 192.4: ‘the clock was and I was late for school’ could be completed with the words slow, missing, broken, melting, buggered; or proposing ‘string’ rather than ‘trunk’ as a way of connecting ‘elephant’ and ‘tree’). Others inviting more creativity can be very easily subverted (ex 200.4: ‘Instead of ……, we went ……’ can be completed with various colourful phrases) though the serious student would be well advised to rein in their creative instincts.

Also, while there are many exercises involving synonyms and homonyms, there are none involving heterographs and heteronyms, and conceptual similarities between homonyms are not explored. Then again, Gregory hasn’t set out or claimed to be exhaustive, and his book is certainly bountiful.

Last time I looked, this book was out of print and only available in quite expensive second hand versions, so when I thought I’d lost it recently (with my copious markups in pencil made over the past few years) I was sufficiently distraught to immediately start work on this long-delayed write up when I found it again. It’s wholeheartedly recommended, both for its presumed purpose as a sourcebook for prepping for SAT, ACT and GMAT assessments, but also as a source of light entertainment. There are also game apps based on many of these exercises (see my Medium article, ‘The Glass Bead Gamenasium’) so I don’t think it’s just me who finds such things playful. Or is it?