The Fruit of the Loom

15 min readAug 30, 2021


It makes sense, literally.

I was fortunate recently to come across the work of Dirk Geeraerts, Professor of Linguistics at KU Leuven in Belgium, and have been especially interested in his work concerned with onomasiology and semasiology.[1] These two long-established disciplines of linguistics study respectively the different ways a given meaning can be expressed, and the different ways a given form can be interpreted, and Professor Geeraerts is at the forefront of innovations in their application and development.

I’m not a linguist, and my interest arises because, for a while now, I’ve been developing games of comparison and variation which play with the different meanings/uses of forms, and the different forms which can express meanings/uses. [2]

Alongside the game, and included as an optional element of play, I’ve developed a framework for classifying these comparisons according to whether their forms or meanings are the same, similar or different. The resulting Loom of Form and Meaning studies the different ways a given meaning/use can be expressed/achieved on the horizontal axis on the Loom (i.e. onomasiology), and the different ways a given form can be interpreted/used on the vertical axis on the Loom (i.e. semasiology), as well as simultaneous variations in both meaning/use and form. [3]

For example, in a game based on a chevron bead, players first choose the object they want to make the subject of the game (i.e. the chevron bead), after which they choose the particular aspect of the object they want to develop (a token of value). The resulting game would then develop from aspects of the physical form and appearance of the bead itself, or else the associated idea of value, or variations which are a combination of both.[4]

The game could also start in other ways. Firstly, players could choose the concept they wanted to develop (e.g. ‘storage’), and secondly choose a form which could represent that concept (e.g. any store or carrier of value, such as a chevron bead). Either way, they will arrive at a form and meaning which together become the subject of the game.

Geeraerts[5] might compare these early stages of pre-play to the referential dimension of onomasiology and semasiology: the process of pairing any form and meaning necessarily involves the formation of a concept (i.e. the conceptual framing of what is to be expressed) and assigning it a name (i.e. the choice of how to express the concept). Not all ways of framing concepts are equally common (i.e. available conceptual categories vary across cultures and individuals and are used with differing frequency, in a manner that Geeraerts calls conceptual frequency), and not all ways of naming concepts are equally common (i.e. available lexical categories vary and are employed with differing frequency, in what Geeraerts calls lexical frequency), and both of these factors work together in the context of real-world experience (experiential frequency) to produce the distinctive modes of expression of an individual speaker within their wider language community.

The main stage of game play is then comparable to what Geeraerts might call the formal dimension of onomasiology and semasiology in which variations of a given form and meaning are studied, in the same way that the initial form and meaning in the game are subjected to variation in the players’ successive moves.[6]

As a result of my encounter with the work of Geeraerts, I have been able to add some useful tools to my descriptive toolbox, so I can attempt a better explanation of how the Loom relates to these linguistic concepts. I’ll do this through the example of how the Loom would distinguish different onomasiological and semasiological relations of fruit across its nine categories.

A1. Etymologically fruit is derived from the Latin fruī which has a wider sense of to enjoy/use, but according to the full Oxford English Dictionary (OED) it enters the written record in English (in 1175) in the sense of ‘vegetable products in general, that are fit to be used as food by men and animals.’ Later the word reacquired a wider sense of anything accruing from action or effort (the earliest example given by the OED is 1300), and also became more specialised in general usage to refer to tree fruit, for example.

I could start my exploration at any point in the conceptual range of the word fruit. I’ve chosen the meaning given in Carl Darling Buck’s Dictionary of Selected Synonyms (1949): ‘a generic name for ‘tree fruit’, covering apples, pears etc.’ This is the point in my illustration that conceptual onomasiological variation is acknowledged, as an initial choice which may continue to be refined as we explore the remainder of the Loom. This choice having been made, formal onomasiological variation then proceeds through A2 and A3 as below.

A2. The OED gives some archaic spellings of fruit which provide variation within the bounds of similarity, and without changing the meaning. Fruct is particularly mentioned as a vestigial English spelling (‘merely a variety’) from the Latin, persisting into the 14th and 15th centuries.

From Buck we can find cognates for fruit in the principle Indo-European languages which are given as synonymous with our target concept, and are similar in form: Italian fruito, Welsh ffrwyth, Danish frugt, Swedish frukt, Dutch vrucht and so on. The typographically identical French fruit is not pronounced the same as English fruit, so is only similar in form overall.

At this point, I am reliant on a prototype-theoretical conception of semantic structure[7]: how could all these categories be truly synonymous when the experiential and conceptual variations must have varied so widely over time and space? But they are close enough to consider them to have the same meaning for our purposes.

A3. Among the other more diverse synonyms given by Buck, we can distinguish some with different etymology but the same secondary semantic associations (e.g. Lithuanian vaisius, with associations of fertility and produce). In such cases there’s an extra thread of semantic equivalence in the motivation which is not shared by synonyms with different motivations (e.g. Greek όπώρα, from ‘late summer’; Sanskrit phala, with associations of bursting), which may nevertheless remain as A3 synonyms as long as the secondary associations do not overwhelm and significantly affect the meaning.

Bomhard’s Nostratic Etymological Index gives three different roots for the concept of fruit: *biry-, which as a verb is enjoy, savor, and as a noun is fruit; *phir-, which has associated meanings related to bringing forth and birth; and *čhana- which also relates to bringing forth, growth, and production. It is the first of these which gives us the Indo-European root *bhrūg-, and the cognates of fruit included in A2, with derivatives of the others as synonyms in A3, alongside those from other languages and families which may or may not share the same associations. (Chinese has one synonym of fruit,guǒ, which shares associations with produce in general. Arabic has ثمرة thamra, which colloquially is a date fruit, but classically means fruit in general, as well as produce and profit.)

B1. All the other common meanings for fruit listed in the OED are related meanings with the same form, including the common compounds, some more entrenched and salient in actual usage than others. There are no ‘true homonyms’ given, for which the OED would allocate a separate main entry.

Also in the B1 category are meanings arising from individual speech acts which would not cause surprise or difficulty, e.g. referring metonymically (category for member) to a banana as fruit, as in: ‘What’s for breakfast?’ ‘Fruit, as usual.’ (Bananas are for breakfast again, as every day. As opposed to: ‘Apples, for a change.’)

Produce is a literal meaning of fruit, and is etymologically earlier than the category of edible vegetable products. Fruit of the loom is an example of the metaphorical B1-type use of this aspect of the word fruit as a source: the target of this metaphor are the muslins and fabrics produced on the loom. (The now well known trade name Fruit of the Loom is a C2-type rhyming pun (loom/womb) on the biblical fruit of the womb, coined in 1856 by Rhode Island textile mill owners Robert and Benjamin Knight.)

B2. This is the category associated with morphological family resemblances, different grammatical forms of the same word: fruity, fruitful. It also includes etymologically related words which are semantically related: fructify, fructose, frugal, brook (in its sense of use, allow, tolerate).

What I allow as similar meaning and similar form in this category is broader than what would be included in the prototype-theoretical conception: some of the words I include here are clearly not fruit per se, but they are somehow related to the original chosen concept and similar in some respects in both form and meaning.

B3. The words apple and orange, on the other hand, are not related to the word fruit except as examples of the generic category. Though the specific word for apple also means fruit generically in some languages (e.g. Latin malum, apple or fruit), it doesn’t in English. Since there’s an overlap (an apple is a kind of fruit) we can say there is a similar meaning, but it’s a completely different form, making such examples B3.

‘Nature’s bounty’ is a metaphor for ‘fruit of the earth.’ Fruit is the target of the metaphor in this case, and generosity/bounty is the source of the metaphor, emphasising in particular the quality of abundance of the earth’s produce.

C1. This category includes less entrenched meanings not everyone would have access to, to the point that they might be surprising or not recognised by everybody — e.g. the use of fruit as (originally US) slang for male homosexual. Other uses which may surprise include referring to a tomato as a fruit, which is technically correct, though in popular understanding it isn’t usually thought of as such, or stalks of rhubarb, which is technically incorrect, but which might popularly be referred to as fruit when baked in a ‘fruit pie,’ though probably not in its raw vegetable form.

C2. This is the category of rhymes and rhyming puns, as well as other imaginatively similar forms with different or unexpected meanings. I had to make up a ‘dad joke’ punning on the word fruit to illustrate this: What do you call a piccolo cut up into pieces and served with sugar and cream? A small flute salad. I tested this joke on a few people (including my own son and my dad), and it was unanimously confirmed as being pleasingly terrible.

C3. And so to the fascinating category of different forms and different meanings. Anything not included elsewhere on the table could be included here, but that’s not very satisfying. I like to include something relevant in this category, though necessarily dissimilar in both form and meaning or else it would not belong here. ‘Rhubarb crumble’ could well be a candidate, given what we said about rhubarb earlier. Or ‘barrenness’ in opposition to fruitfulness. Or the word used by St Jerome when translating the generic term for fruit in Hebrew (פְרִי peri) as malus in Latin, but with its homonymic meaning of wicked which Jerome deliberately punned on.


This extended example on fruit is an unusual (for me) and specific application to linguistics which is narrow in that all examples are limited to words, but wide in that all languages and proto-languages may be included. I chose to make pronunciation a sufficiently significant component of form to differentiate French fruit as A2. Putting French fruit at A1 instead of the English wouldn’t change the top row much, other than swapping the English and French terms around. Using Littré then, instead of OED, might give B1 a different flavour, but metonymical (category for member) and metaphorical (with fruit as source) uses should carry across in translation, and the set of family resemblances in B2 would need to include related words in French, as indeed they may also have been included even with English fruit at A1. We’d need an equivalent to ‘nature’s bounty’ in French as an example of metaphorical use of fruit as target in B3, and while some less entrenched usages in English would be nudged to C2 (e.g. the US slang), some others in C1 relating to surprising categorisation (e.g. referring to rhubarb) would still carry across. I’d need a new dad joke in C2, since flute/fruit doesn’t rhyme in French, but C3 needn’t change much. French pomme for apple would be B3, along with English apple — they’re completely different word forms from fruit, with overlapping meaning.

The same applies in the case of the vertical axis. Putting the English meaning for fruit of product/result in A1 would relegate to B1 its meaning as a generic term for apples, pears and such, but would not affect much else in the first column, and would still allow its US slang use to be included in C1, for example.

Putting this C1-type US slang use as the meaning in A1 would have more impact on the Loom as a whole, especially the top row of synonyms, just as putting a more exotic A3-type form such as Sanskrit phala with its associations of bursting would have a wider effect, especially in the first column of homonyms.

The way I’ve been using the Loom until now, there’s not only no restriction but also not really any concept of successive steps across the Loom. I tend to relate things directly back to A1 usually. Barrenness in C3 doesn’t need to go via fruitfulness in B2, but it more neatly aligns the words grammatically if it does. That said, the path across the Loom is sometimes interesting. An example in French from Pierre Guiraud, supplied by Geeraerts[8], of chat/cat in A1 going via chas/starch in C2 (or C1 if you follow only pronunciation and not spelling), and via maroufle/cat in A3, to meet again in C3 at maroufle/starch is very satisfying.

I guess there could be a separate matrix for each of English and French, for example, even limiting further the period/textual base from which examples are drawn, with etymological cognates and translations as associations across matrices, but for my purpose I value the richness a wider multi-lingual approach brings. What I’m trying to find by using the Loom are interesting perspectives on whatever idea is in A1. For example, it’s particularly interesting to look at different motivations for synonyms of the same concept in other languages. It tells us something more about how (or even if) other cultures perceive the same concept, and potentially gives us access to a wider range of novel figurative uses and interpretations.

I wouldn’t normally limit my examples to language, let alone a single language, but it would certainly be possible to apply the Loom in that way. I’ve experimented with ‘faceted’[9] games which overlay a particular geographical or historical context (e.g. sub-Saharan Africa pre-1500 CE) or modality (e.g. visual arts) to govern the choice of examples used in the Loom, or any game of comparisons derived from it.

Typically, having chosen ‘fruit’ at A1 in a game ‘faceted’ on the arts, I might introduce at C2 an Arcimboldo painting in which a selection of fruit weirdly resembles a face, or a vanitas still life painting from 16th/17th century Flanders where rotten fruit is the source (B1) of a metaphor for decay and death, or the artist Guillermo Galindo’s absurd idea (C3) of a tennis shoe found on the moon which in turn parallels the musician Karlheinz Stockhausen’s example of fruit – in his case an apple – found on the moon.

If Wittgenstein’s aim (in Philosophical Investigations, 1953) is to adequately define a category such as ‘game’ or ‘number,’ Baldinger’s (in Baldinger, K., Semantic Theory, 1980) to identify a whole which encompasses a multiplicity of expressions, and Husserl’s (in Experience and Judgment, 1939) to intuit the essence of a fact or quality such as ‘red’ and reduce reality to idea by removing difference from its variations, my use of the Loom is inversely to multiply the particular cases in which something can be found as a variation across cultural, linguistic and temporal horizons and contexts, to be more aware of the ways we all experience, create and connect with the multifarious heterogeneity of reality, and enable ourselves to face each other in constructive play to generate together these potentialities in a way that is stimulating and fun.

Is that nuts?

I’m grateful to Professor Dirk Geeraerts for giving me several valuable pointers in his own work and the wider literature, for providing the impetus for the application of the Loom of Form and Meaning to the concept of fruit, and for supplying the subtitle for this article. Any mistakes and misrepresentations of his views and others are entirely my own responsibility.


[1] The article of Professor Geeraerts which first attracted my interest was Four guidelines for diachronic metaphor research, in Diaz-Vera, J.E. (ed), Metaphor and Metonymy across Time and Cultures, 2015. There Geeraerts explores the metaphor ANGER (TARGET) is HEAT (SOURCE) by analysing other meanings of HEAT (i.e. NOT-TARGET is SOURCE: aside from things being literally hot, figuratively heat also means sexy, stolen, funky, jazzy &c), and the frequency of different synonyms for ANGER (i.e. TARGET is NOT-SOURCE: in a body of old English texts, for example, use of expressions of heat to represent anger are shown to be marginal, and literal expressions dominate, such as ire, wrath &c). Having read the article, I wondered about the unexplored combination of NOT-TARGET is NOT-SOURCE, and in particular whether certain examples of this type, in particular antonyms of the TARGET and SOURCE, might be of interest in the overall analysis of the original pattern. The question is already implicit in some of the Old English words listed by Geeraerts as synonyms of anger, such as unblide, which is the negation of blide (blithe), and as far as I know is unrelated to heat: how important is heat (or its opposite) as a SOURCE when the TARGET is the opposite of these words for anger. In some cases it may be consistent (e.g. calm/cool), but not always (e.g. pleasure/hot), showing that the heat spectrum as a SOURCE may code more generally for intensity of emotion/feeling, rather than anger in particular.

[2] Such a game may start with any object or concept. A game beginning with a chevron glass bead, for example, may use the meaning of ‘value’ taken from its historical use as a token of exchange in early cross-cultural contact by European global explorers and settlers, and compare this to the use of these beads as a valuable grave good for some cultures, for trading in the afterlife. Or else it could build on visual associations with the bead’s characteristic 12-pointed radial design (e.g. comparing to the timber frame roof of a pavilion in the Great Mosque at Xi’an, a rose window in the Cathedral at Reims, a radiolaria skeleton, a chlorophyll molecule), or the zigzag chevron pattern which gives the bead its familiar name in English (e.g. comparing to the trill notation or a triangle wave in music, a mitre joint in woodworking, or a twill pattern in weaving).


In my most recent card game, the Glass Bead Game Variations, the 18 cards featuring different glass beads have each been assigned a meaning. Having chosen the bead card which will be the subject of the game, players are asked to: ‘Reflect on the ways in which the bead overflows or does not fully fit its assigned meaning. Also, look at the bead carefully and think about what else it might normally mean to you.’ This stage of pre-play sets an initial course for what the subject and range of the game might be, and potentially what it is not and should not be, by arriving at an initial form and meaning which will be the starting point of the game. The players can then continue: ‘With all this in mind, imagine what the bead might represent to you in other subject areas drawn from the context cards.’ This is the main stage of the game, where the players make their moves by reinterpreting the form or reexpressing its meaning in different contexts drawn from the 36 context cards in the remainder of the deck. In a final stage, players may holistically consider the movement of the game from beginning to end, and how the individual moves complement each other and cohere as a whole.

[4] For an example of this as an actual game by two players, see

[5] Geeraerts, D., Entrenchment as onomasiological salience, 2016. In Schmid, Hans-Joerg, Entrenchment and the psychology of learning: how we reorganize and adapt linguistic knowledge, 153–174. Berlin: De Gruyter Mouton.

[6] A further layer of context sensitivity is added to the moves of the game as aspects of the reinterpretation or reexpression must be linked by players to one of the contexts currently in play. A given variation may be legitimately linked to any of several contexts: ‘A phenomenon is classed according to the context or discipline in which it is considered (so that its various aspects occur at many different places). For example, ‘coal’ has no single place in UDC. The petrological aspect is at 552.574, … the mining aspect at 622.33, … more peripheral aspects are at still other numbers.’ BSI, Universal Decimal Classification, 2005. In fact, the Universal Decimal Classification provides the basis for the 9 knowledge-based cards of the 36 context cards, the other cards representing types of uncertainty (8 cards based on a system set out in Savin-Baden, M., Howell Major, C., New Approaches to Qualitative Research, 2010), physical/sensual awareness (7 cards based on the chakras and their popular associations), and emotional intelligence (8 cards based on the Lovheim cube of emotions in Lovheim, H., A new three-dimensional model for emotions and monoamine neurotransmitters, 2012, and 4 based on the domains of emotional intelligence in Goleman, D., Boyatzis, R. & McKee, A., Primal Leadership: Realizing the Importance of Emotional Intelligence, 2002).

[7] Wittgenstein, L., Philosophical Investigations, 1953

[8] Geeraerts, D., Theories of Lexical Semantics, 2010

[9] A term from library and information science referring to a semantic ordering which is based on an aspect of the information in hand, e.g. period, time, place.