Setting Things Right: The Manifezto Pt.6

26 min read · Sat 25 Apr 2026

Hi all, welcome back to Setting Things Right, aka The Manifezto. Quite a long one this time, sorry – there's a lot to think about! The usual caveat: these are just my ideas and opinions – please do add your own thoughts and disagreements in the comments!

Part 6: Putting Things Together

In the previous post, I looked at various ways to modify fodder and tokens within a 'definition & wordplay' clue. These usually involved rearrangements – either disorderly as anagrams, or in a more orderly way by reversal, cycling or some other well-defined movement of the letters involved - and selections, in which only some of the letters are chosen. Although concentrating on these particular devices, the key takeaway was really to pay attention to cryptic grammar, ensuring the clue as a whole provides a grammatically correct representation, description or instruction alongside its definition. Otherwise you'll be drawing a Duck-Rabtib instead of a Duck-Rabbit.

This time, we'll consider the remaining possible elements of a 'definition & wordplay' clue: constructors, qualifiers and links. And again, rather than provide an exhaustive list of specific words for these, the focus will be on how to use them in a way that maintains cryptic grammar.

Constructors

I'm using the term constructor to refer to those parts of a clue that tell the solver how to put different elements together. It's deliberately a rather cheeky piece of jargon, as in the US 'constructor' is often the name given to a puzzle writer (in the UK generally called a 'setter') ... so partly a bit of a tease for our American friends. But anyway, the modifiers we looked at last time act on individual 'building blocks' (your fodder, or tokens resolved into fodder) to make changes to them, preparing them for use, whilst constructors then put two or more of them together in various ways. What sorts of ways?

Unindicated charades

Charades, just like in the parlour game, break the answer up into smaller chunks and present them together. The simplest way is not to use an indicator at all, but just to put the elements side-by-side. This might often seem to be the 'cleanest' construction. Problems start to arise, though, when you introduce 'elements' that are more than simply synonyms, abbreviations or similar, that readily resolve to a 'thing'. If you're using some 'active' wordplay to generate that element ("Doctor Strange", say, for an anagram), or if you have some sort of descriptive phrase ("it produces seeds" for EAR, referring to wheat perhaps), it might not sit comfortably in a simple charade:

Police searches finally found Doctor Strange (8)

Here we have a definition, "police searches", with a charade of D ("finally founD") and RAGNETS ("doctor STRANGE"). You're being asked to just plonk those two elements together. And whilst it's not the most elegant clue, I think it's probably simple enough to get away with that structure, in which the 'passive' D and 'active' RAGNETS are mixed.

By the end of autumn it produces seeds (4)

Here we have a very similar construction. The definition is "by"; then there's a charade of the two elements N ("the end of autumN") and EAR ("it produces seeds"). Even though the construction is pretty much the same, in this case I'd say the clash between the 'passive' N and the 'active' EAR is too jarring - it no longer feels like a 'natural' list of things. That's compounded by a rather sneaky definition, too. As is often the case, it would be futile to try to frame this sort of nuance in terms of rules – no rules! – but instead to use your (and your test solvers') judgement of what is accurate, fair and (particularly in this case) stylish.

Juxtapositions

Juxtapositions are ways to indicate that two elements go together side-by-side, and can be helpful for the surface reading, or to specify a particular ordering of elements, when constructing charades. They might not imply any particular order – X could be "and", "with", or "neighbouring" Y, say, for either X+Y or Y+X – or they might explicitly indicate that one element goes before or after the other – "before", "prior to", "introducing", say, or "after", "follows", "heralded by", etc.

Some juxtaposition indicators only work with a clue that's either Across or Down. For example, if X is "supporting" or "held up by" Y, that'll only make sense if the answer is being entered in a Down light in the grid, whilst if X is "facing" or "alongside" Y it only really suggests an Across answer. Often it's clear when this is the case, but there are some indicators that seem to be a little controversial, so it's worth looking at specific examples – although these are by no means intended to be exhaustive, just examples to illustrate the sorts of consideration to be taken into account when selecting a juxtaposition.

"X on Y" in a Down clue, to me, clearly suggests that X is before, on top of, Y. There's an argument that it could be the other way round too – if a fly is "on" the ceiling, it's actually below it. (Although wouldn't that really give you, reading down, CEILINGYLF? So maybe "smoke detector" is a better example!) Personally, I think using "on" in that way in a Down clue is poor stylistically, as it so much more naturally reads as being "on top", but you can't really deny that it's accurate ... to complete the three principles, I'll let you decide on whether it counts as fair.
"X on Y" in an Across clue, on the other hand, doesn't really imply any specific ordering. In fact, "X on Y" in an Across clue is often preferred to indicate Y+X, as it suggests that X is being "added on to" Y. I also prefer this interpretation, but I don't think the alternative X+Y feels particularly unnatural or tricksy, so personally I'd be happy to see it being used either way.
"X to Y" is often disliked, or even not accepted. I don't think there's really a strong objection – Chambers, for instance, defines "to" as "beside", "near", "in contact with". I guess the objection is that it perhaps doesn't feel completely natural, no matter what the dictionary says. And that is, after all, a reason I've given for avoiding "X on Y" for Y+X in a Down clue, so perhaps it's a little inconsistent, but to me this feels more justified. I'd tend to avoid for relatively simple puzzles (of course, Fez and Cranberry puzzles don't usually fall into that category, so I'm usually happy with to) and because, to me, the most 'natural' interpretation is "beside", I reserve to for juxtapositions in Across clues only.
"X having Y" or "X has Y" (or, in a contracted form, "X's Y") are often seen for X+Y, and I think fair enough, suggesting Y is being given or joined to X perhaps. But to me, this seems a bit vague and less justifiable via the dictionary – Chambers' first four definitions are "to hold", "to keep", "to possess" and "to own", all of which are more strongly suggestive of containment. And yet, as a container (which we'll come on to next), "having" or "has" also feel a bit too vague, when there are specific verbs (such as those four just listed from Chambers) that will almost always do that job in a clearer way. So, whilst I wouldn't raise an objection, personally I prefer to avoid "having" or "has" for either usage. But it is is often very tempting (especially that contracted form, X's Y) and it's one case where I'll sometimes resort to using it for the surface and rely on that get-out-of-jail-free concept 'precedence' to justify it to myself.
"X by Y" doesn't seem particularly controversial, so why include it here? I often see "by" used in a Down clue and find it a bit jarring: I think the natural interpretation is "at the side of", and so for me "by" only really works for juxtaposition in an Across clue. Of course, you could justify it as simply "near", which would be OK in a Down clue, but for me that's just not great stylistically. I think it's similar to "neighbouring", which I'd say is more clearly one to restrict to Across clues only.

Perhaps the key point here is that, in selecting a juxtaposition, you shouldn't be looking to find if it's 'allowed' in some book of 'crossword rules' but instead you should be using your judgement as to whether it clearly and naturally does the job you intend, and maintains cryptic grammar across the whole clue. And if you are happy that it does, then you've made a good choice, whatever that supposed 'rule book' might say. That applies equally to your choice of all indicators, not just juxtapositions. I may have mentioned previously: no rules! Maybe that should be extended to: no rules, only reasoning!

Before moving on, just a quick note to acknowledge that here I've only used X and Y to represent elements. Your choice of juxtaposition may well be influenced by the type of element those X and Y are – because, as ever, you need to ensure that the cryptic grammar is accurate. You can't, for example (and ignoring the clunky surface for now), have "Police searches daughter next to Doctor Strange" for those DRAGNETS, because the 'active' instruction "Doctor Strange" isn't a concrete 'thing' that you could put 'next to' something else. But you could conceivably have "Police searches Doctor Strange next to daughter", because now you're asking the solver, in one instruction, to make an anagram of STRANGE beside the letter D.

Containers and contents

Putting one element inside another is a favourite device for many setters – it does somehow seem to throw up many of the best wordplay opportunities! You might want to indicate either containment (X is put around Y) or insertion (X is put inside Y), and as with all types of indicator you can find plenty of examples here at MyCrossword or over at the Clue Clinic.

Anything that suggests either containment or insertion, as required, could potentially be used: as ever, the key is to ensure the way that you use your indicator maintains the cryptic grammar not just within the 'container and contents' element but across the whole clue. So, if you have some 'active' wordplay – "Doctor Strange" for the string RAGNETS, say – you need to make sure you don't then treat that element as if it were a noun, by trying to put it in or around something. A couple of quick examples to illustrate how to maintain cryptic grammar:

Attractive bar staff enjoy good time after venue's closing (6)

Here we have "enjoy" (Chambers: "to possess [with delight]" or "to have the use or benefit of") used to indicate insertion. The parse is:

Definition: "attractive bar" / MAN ("staff", as a verb) containing ("enjoy") G (abbreviation for "Good"), then T ("time") following E ("venuE's closing", with "closing" as a noun suggesting the final letter).

Even though on the surface that "enjoy" seems like the right tense (because the surface "staff" is a plural), for the cryptic reading it doesn't work: "staff" here is just a token that resolves to the set of letters MAN. In this case, the solution is simple: "enjoys" would work in the cryptic reading but maybe a little odd for the surface, but "enjoying" does a good job in both the surface and cryptic reading. So, we have a Duck-Rabbit, yay!

Attractive sort of French artist embraces German (6)

Now the definition is "attractive sort", which is perhaps a bit of a stretch ("he's a magnet for the ladies"?), but the key point is the cryptic grammar. The parse is:

"Attractive sort"; definition "of" wordplay: MANET ("French artist") contains ("embraces") G ("German")

But the wordplay (in its entirety) is an 'active' phrase, so it jars with the linking word "of". Again there's a simple solution: using "embracing" enables the wordplay to be read as a noun, a 'thing', so the link no longer feels jarring. We'll look in more detail at link words shortly.

Deletions

We've already seen reductions as modifiers that remove parts of a single wordplay element. In a deletion you're removing one element from another. Indicators can be classified as either departures (or subtractions, where X is removed from Y: "X cancelled by Y", say) or expulsions (ejections, where X gets rid of Y: "X cancelling Y"). And again, there are loads of examples here at MyCrossword or over at the Clue Clinic – in both cases, the lists include not only deletions (removing one element from another), but reductions (modifying a single element).

And there's no need here to trawl through further examples. When selecting a deletion indicator, the principles remain the same as for all indicators: is it accurate, is it fair, and is it stylish? Plus, of course, does it work within the clue as a whole? So, pay attention to cryptic grammar!

Intermingling

One further way of putting elements together is by 'intermingling' – this might involve sharing parts of each element, peppering one element throughout another, or maybe devices like Carrolisms in which a whimsical portmanteau word is created from two elements. These constructions aren't seen very often, and I won't spend a great deal of time on them here (not least because the important issues remain just the same: accuracy, fairness, style and cryptic grammar), but just as a quick example here's a lovely clue from Coot's Featured puzzle #7 on MyCrossword:

Public transport sufficient with only one morning crush (7)

The parse is:

TRAM ("public transport") together with AMPLE ("sufficient"), but sharing the AM part of each word ("with only one morning"); definition: "crush"

Qualifiers

Qualifiers are used when you need to provide a little extra information about your definition, or about your tokens, in order to make them fair to solvers.

'Unusual' words

These might include words that are obsolete or archaic, dialect, foreign or non-standard English. To be fair to solvers, you need to indicate if you are using a term that is American or Scottish, is only used in a particular regional dialect, is now considered obsolete or archaic, or is only seen in 'txt-spk', say. So you can qualify such words with all sorts of indicators such as: "Trump's" for an Americanism; "in Perth" for either Scottish or Australian, "locally" for a dialect word, "old-fashioned" for an archaic word, "on phone" for txt-spk, and so on.

Aural wordplay

Homophones are usually presented as a type of wordplay, but really I think they are just 'qualified' elements. "Pet animals (4)" doesn't get you either DEAR or DEER, but if you qualify either of those elements with a homophone indicator then you've got a clue: "Pet animals in the auditorium (4)" for DEAR, maybe, or "Caught pet animals" for DEER.

As crosswords are just whimsical, fun puzzles, I don't think strict homophones are required: they might be better thought of as 'soundalikes' or 'puns'. A little bit of stretchiness in the aural similarity can be nicely amusing and effective. But there's a limit to this: if your homophone relies heavily on using a particular accent (be that Cockney, or Scottish, or Australian, or American, or whatever) then it ought to be indicated as such.

Of course, what classifies as a 'normal' pronunciation will vary between individuals. Chambers and other sources provide phonetic spellings that can help out if there's any doubt. One objection sometimes seen is that a word doesn't resemble its intended target for 'rhotic' speakers: say, equating CAR with the spirit KA, for a rhotic speaker won't work because the ending R sound isn't dropped. But to me, that's a weak objection: although a rhotic speaker would pronounce the R themselves, they would surely recognise and understand the word if hearing someone else talk about their "cah"? Complaining that "it doesn't sound quite that way to me" seems a rather curmudgeonly and self-centred quibble.

One key point with aural wordplay is that the soundalike must retain that property in the final answer. You could use a homophone to indicate part of an answer, but nobody really knows what a split or reversed word sounds like. So, if you mess around with the homophone in some way you are destroying its homophonic nature, and thus making your clue inaccurate and unfair. The exception is where a homophone refers to a single letter: "eye" for I, "you" for U, "sea" for C, and so on.

And as with other indicators, rather than trawl through a list here, I'll refer you to those at MyCrossword and the Clue Clinic.

Spoonerisms are a particular type of aural wordplay, relying on the supposed habit of the Reverend William Archibald Spooner of mixing up the opening sounds of words in phrases: "catflap" instead of "flat cap", say: "A small opening in Spooner's flat cap (7)". It's not a favourite for many solvers, perhaps because it feels archaic and relies a little on 'insider knowledge' of crosswords (where else would you see it?), but used sparingly it can be fun and effective.

There really is only one way to indicate a Spoonerism, and that's by reference to Spooner himself: "as Spooner might say", "Spooner's", "according to Spooner", etc. Sometimes you'll see "the Reverend" instead, but for me that's unfair: a crosswording 'newbie' might be puzzled by a reference to Spooner, but could at least look it up and discover what the setter means; but a reference to "the Reverend" or "absent-minded don" absolutely requires that you have that cruciverbal 'inside knowledge'. And a final point on Spoonerisms: they rarely work with short words. Think about whether the Spoonerised phrase is really one that he might have conceivably, naturally, have said and gotten mixed up.

Definitions by example

Generally, if the answer is the name of some 'higher' set that includes your definition or token, then that definition or token needs qualification as an example: "setter, perhaps" for DOG. But if it's the other way round – your definition or token give the more generalised set – then it's fair, and indeed more accurate, not to give a qualification: SETTER can be defined as "dog"; it's not "dog, perhaps", as a setter clearly is a dog (whereas a dog is not necessarily a setter). Other ways to indicate an example include "say", "for example", "for one", and so on, or just a question mark: "setter?"

One common 'error' here is to imply that a descriptive word provides an example of a thing.

Mainly why drink is blue? (5)

Here we have WH (most of, or "mainly", WH[Y]) and ALE ("drink"), that wordplay being ("is") "blue" as an example (hence the question mark). But "blue" isn't an example of a WHALE. OK, if you're asked to name some examples of whales, you might reel off "blue, killer, sperm ..." but that's in the very specific context where "whale" has already been mentioned and is thus implied to form part of your answer. Asked to name examples of marine creatures, you wouldn't say "squid, turtle, blue ...". Instead, it is a blue whale in full that provides the example. An orca or a narwhal provide examples of whales, but a "blue" or a "pilot" on their own don't. At a stretch, you might get away with "It might be blue" (giving you a descriptive phrase rather than a 'thing', so needing care with the cryptic grammar) but personally I find that a bit lacking, style-wise.

Names

There's a debate as to whether people's names should have 'definition by example' indicators. So, is it fair to define MCCARTNEY as simply "Paul", say? With names, things get a bit trickier to analyse.

Is "Paul" really an example of a "McCartney"? Or is "McCartney" an example of a "Paul"? Well, it's probably both ... in which case, surely you can choose to go without a definition by example?

Personally, I'm not sure that's entirely fair. And perhaps it's better not to think of names in terms of "definition by example" – people aren't really examples from some supposed 'name set', and any such sets would be difficult to conceptualise, as each is somehow a subset of the other. I'd prefer to simply think of these as "definition by name", and then the 'problem' vanishes. That is, you don't need to think if a surname is an example of a forename or vice versa, but instead concentrate on how it might be fair to clue each individual name.

So, for Paul McCartney, I'd say:

McCartney isn't that common as a surname, at least as one that can be fairly clued by reference to some well-known person. It could be Paul; it could also be Linda or Stella, perhaps, but that's probably it. So if you want to clue PAUL, then "McCartney" on its own feels reasonably fair. "McCartney, perhaps" doesn't really help the solver any more, so I don't think it's absolutely necessary – but to me it does still feel a little odd without.
Paul, though, is a pretty common first name even amongst celebrities. It could be McCartney, but it could just as well be Simon, Young, Mescal, Revere, Newman, Gascoigne ... and so on. So if you want to clue MCCARTNEY, just "Paul" on its own feels rather unfair. "Paul, perhaps" doesn't narrow down the options, but it does feel like a fairer, more accurate, definition – so I'd say that's OK. "Musician Paul" or similar would be even clearer, and for me that would usually be the way to go, to ensure fairness.

But it doesn't always work that way round:

Smith is a far more common surname than McCartney, so in this case if you want to clue WILL or SAM or PATTI or ADAM ... well, you get the idea – "Smith" on its own is just too vague. So again, whilst "Smith, perhaps" doesn't narrow down the options it at least feels fairer and more justifiable; "Scottish economist Smith" gives a clearer definition. And furthermore, as it's a first name, it might still feel a bit odd to clue it simply as that – if asked to list economists, you're not going to say "Adam, John, Milton ..." but "Smith, Keynes, Friedman ...". So it might be even better to include a suggestion of that: something like "familiarly" or "to his friends", perhaps?
Frasier isn't at all common. It can only reasonably refer to Frasier Crane from the sitcom; cluing CRANE as "Frasier, perhaps" seems unnecessary – what does that "perhaps" really add to help the solver? If it's "Frasier", it pretty much has to be CRANE.

So when it comes to names, I don't think an indicator is always strictly needed, but (i) it will very rarely be inaccurate or unfair to use an indicator, and often feels like a more natural reference, so something like a 'definition by example' indicator is a safe option if in any doubt; and (ii) if some indicator really is deemed necessary, it's usually clearer – and so fairer for the solver – to provide something that describes the person, rather than just a 'definition by example'.

Links

Link words join your definition to its wordplay. Often a link word isn't necessary, and indeed many setters prefer to avoid link words if at all possible, but a link can sometimes be useful to ensure a coherent surface. Personally, I think link words are sometimes unfairly maligned; if used properly they maintain complete accuracy within the clue and can help to provide a more pleasing surface, so why discount them? And in some cases, especially when using 'active' wordplay, a link provides a much more natural cryptic reading too:

Doctor Strange to provide precious stones? (6)

Here, you're asking the solver to perform an operation (anagram the set of letters STRANGE) in order to get something meaning "precious stones". With that 'active' instruction, it seems far more natural, to me, to include a link – "Do this in order to get this" – rather than just plonk the definition beside the instruction – "Do this: this" – which feels a tad jarring in mixing an active instruction with a 'passive' definition. So, whilst I agree that avoiding link words can often lead to 'better' clues (for one thing, you don't signal the break between wordplay and definition so explicitly), that's not always the case.

Anyway, where no link is used, the definition can be read a bit like a caption. Really, it ought to be divided from the wordplay by a colon or semi-colon, say:

Big American who fights soldier? (5)

Here, the definition is "big", given by GI ("American who fights") alongside ANT ("soldier", strictly as an example?) Really this should be read as:

Big: American who fights, soldier? (5)

... with that colon serving to show that "big" is a sort of 'caption' for the 'picture' drawn by the wordplay (and the comma clarifying that the wordplay is in two pieces). But that punctuation can fairly be omitted, leaving the solver to work out where the 'breaks' are meant to appear – indeed, spotting those breaks is one of the joys of solving! I'll look at punctuation in more detail next time; for now, suffice to say that omitted punctuation is often (but not always) fair, presenting the clue in a terse 'telegram-ese' manner but not essentially changing the intended meaning. Added punctuation is more often (but again, not always) unfair, as it might materially affect the grammar of the clue so that you no longer 'say what you mean'. (I've sometimes heard setters and solvers say that all punctuation can just be ignored, but – and here I can be blunt, as it affects accuracy as well as fairness and style – that is simply not true.)

Direction of links

Some links are non-directional, for example you can present the definition and the wordplay, just saying that the answer is given by both of them. Or, the definition or the wordplay, saying you could think of the answer as being either of them.

Other links are clearly uni-directional, with one part leading to the other. Here, you need to make sure that it is the wordplay that leads the solver to the answer. You might provide wordplay for a definition, say, or obtain a definition from the wordplay. If you think of it in terms of an arrow pointing from one part to the other, you must ensure that the arrow always points from the wordplay to the definition: wordplay -> definition or definition <- wordplay. Otherwise, you're asking the solver to find some wordplay given a definition, which I'd say is a completely unnatural way to think about a clue: you want the solver to be finding the definition!

Other links can be seen as bi-directional. You can imply that the definition is the wordplay – that is, they are the same thing – or equally imply that the wordplay is the definition. You might present the definition as some wordplay, or have the wordplay as (that is, being) the definition. But that's not quite the same as 'non-directional', as you are still implying that one part leads to the other – it's just that the implication could be read in either direction. In practice, this distinction isn't often important, but one consequence is in using it as part of a double definition:

"Striker is one of a pair (5)"

Here we have two definitions for MATCH, "striker" and "one of a pair". But the implication is that only one of these is the definition, which is equivalent to some wordplay. Many would ignore this subtle distinction, but I do always find it a little jarring. Instead of X = Y (a truly non-directional link), you're presenting either X -> Y or X <- Y, and whichever one of those you decide to interpret it as, there's a definition at the wrong end of the arrow. That said, whilst I personally prefer to avoid it, I think it is usually probably OK and wouldn't raise a quibble ... you could argue that you're presenting X <-> Y, with both definitions at the right end of the arrow.

A few particular examples of link words that I think are worth considering:

Wordplay to definition. That "to" seems to suggest that the wordplay leads you to the definition, but it doesn't actually say that: there's something missing. Instead you need something like "to get you", "to provide", etc. (Of course, if your answer is a verb, then you can define it as "to play", "to drink", "to run" ... so the "to" is no longer a link but part of the definition.)
Wordplay but definition is one I've encountered on occasion, and it doesn't really make sense to me: "but" doesn't imply one part leads to the other, or that they are the same .. so for me, it simply doesn't work. Maybe "but also" would do the job?
Wordplay with definition. I've often seen objections to "with" as a link word, but personally I don't see any problem here. It's similar to "having", perhaps ... "here's a definition that has as its elements this wordplay". To back that up, Chambers defines "with" as "possessing", and also as "using" and "featuring", all of which, to me, provide ample justification for "with" as a link from wordplay to definition, as well as the more generalised Chambers definition as "denoting ... association or connection".
Definition with wordplay. Even more controversial, maybe, is the use of "with" the other way round. That generalised Chambers definition of "with", I think, is sufficient to justify its use, but there's also the more explicit Chambers definition "by means of" that I'd say works perfectly as a link this way round. Or, more whimsically, you might have "Here's Michael Fish with the weather", meaning "bringing you". So "with", for me, works nicely as a link word. Note that I'm not relying on the meaning of "with" as "beside" (here's the definition, and next to it the wordplay) – I think it's clear that "Wordplay next to definition" doesn't work at all, and that's what that interpretation would amount to: so whilst I happily accept "with" as a link, it's bi-directional rather than non-directional ... with the usual consequence for double definitions.

Unexpectedly placed definitions

Whilst the definition is almost always found at the start or end of the clue, that's only because in most cases it's necessary in order to ensure correct cryptic grammar, with that definition either acting like a caption (no link) or providing one end of a pair of linked pieces (the other piece being the wordplay). There's no rule that the definition has to be at one end. A few examples where you might see this:

A link word might be placed at one end of the clue, rather than between the definition and wordplay, so instead of "definition from wordplay" you might have "for definition, wordplay" which is still accurate from the point of view of cryptic grammar:
- For believers, the first Sabbath (7)
- (THEISTS: For THEISTS, THE + IST + S)
A 'link' might work in a split way; rather than "definition by means of this wordplay", you could have a more active "get the definition by doing this wordplay operation":
- Get awards by moving across (6)
- (OSCARS: Get OSCARS by anagramming ACROSS)
Or, you might be able to refer to the definition within some wordplay in a way that still ensures the intention or focus of the clue is clearly on the definition (most often using "this definition" to emphasise that's the element the solver's looking for):
- Mixing this writer up with a lot of others might get you in hot water (5)
- (TWAIN: Mixing TWAIN up with OTHER(S) might provide: IN HOT WATER)

Summary

Whenever you use a modifier, constructor, qualifier or link, you're always introducing some further complexity to the clue. And so, you need to take extra care to ensure that the cryptic grammar still works across the clue as a whole
Lists of indicators are available here at MyCrossword and the Clue Clinic ... these are incredibly helpful for reference, but always think about how you are using them. Just because an indicator appears in a list, it doesn't mean it can be used in just any old way: use your judgement to decide if the indicator really works, and check that the way you use it maintains the overall cryptic grammar
So, your judgement (and that of your test solvers) is the best means of deciding if any indicator is accurate, fair and stylish; there isn't a 'rule book' to consult, and there isn't any need for one. No rules, only reasoning!
Did I mention? Cryptic grammar!!!

Next time we'll wrap up this examination of clues with a general recap and overview, and a look at a few loose ends, for example:

Looking at the scope of indicators: how and when can an indicator apply to multiple tokens or bits of fodder? (E.g., in "first of all, we are Yorkshiremen", the scope of that "first" is surely just the "all", so it can't lead to AWAY ... or is the indicator "first of all", specifying the 'scope' as being all of the following words to get WAY?)
Use of punctuation, capitalisation and spacing: when is it fair to omit or introduce punctuation marks; is 'lift and separate' fair (that is, omitting a space, e.g. "Treasure inside a room (4)" being read as "in / siDE A Room", with the hidden indicator "in" attached to the fodder SIDEAROOM); when can you capitalise or decapitalise words fairly?
A couple of devices not yet really considered: &lit clues, and reverse engineering
And, if time permits, a re-visit to the 'Ximenean vs Libertarian' debate, with some thoughts on how 'libertarian' you can be without straying into 'inaccurate' or 'unfair' territory ... and on how 'Ximenean' you can be without stifling the fun and creativity at the heart of puzzles

Til then, cheers!

Fez

0 comments

Be the first to comment!