Please use the comments form attached to this page for any questions, thoughts, ideas etc about cryptic crosswords to which you would like a doctor to respond. Biographical information about the doctors on call can be found on the Meet the Team page.
The Site Feedback page should now be used only for comments about the site itself.
The most recent comment threads are shown below; to view previous threads, select the ‘Older Comments’ link at the bottom of each page. Individual threads will not be split across pages.
Hi. a couple of questions that I was wondering about. Can you use a word like ZOOT (on its own) when it only appears in Chambers as part of ZOOT SUIT? And can you pluralise a word like RAW when Chambers gives the noun definition as (with ‘the’)? If so, how would you go about it?
Hi Alex, and thanks for two very good questions.
1. If a word appears in English dictionaries only as part of a multi-word expression, such as ‘zoot’ in ‘zoot suit’, ‘gnamma’ in ‘gnamma hole’, or ‘pe-tsai’ in ‘pe-tsai cabbage’, then it cannot feature as the answer to a crossword clue. This is basically because the word has no meaning in the English language. The exceptions to this rule are proper nouns, such as ‘Petri’ in ‘Petri Dish’ or ‘Bronx’ in ‘Bronx cheer’; these are valid as single-word entries, although of course they must be defined with reference to the person, place etc rather than directly to the compound (ie ‘dish’ would not be valid for ‘Petri’, but ‘German bacteriologist’ or ‘German dish producer’ would be).
2. Where Chambers shows a word, or a sense of that word, as ‘with the‘, as in the nouns ‘raw’ and ‘occult, the only way that these can appear as an answer is in their complete form as shown in the dictionary, ie ‘the raw’ and ‘the occult’, defined in line with the dictionary definition; this is because they are (uncountable) idiomatic expressions, and as such they have no plurals, and they cannot be used alone with the meanings ascribed to the ‘with the‘ forms. Similarly, ‘the pits’ is a valid answer, but ‘the pit’ is not. Much the same applies to any word, or meaning of a word, shown as ‘with xxx‘ – the ‘summon’ sense of ‘drum’ is qualified by (‘with up‘), so ‘drum up’, defined by ‘obtain through vigorous effort’ or similar, would be a valid answer (and of course the various inflections of the compound would also be allowed). It’s tacitly accepted that another compound verb which shares the same preposition can define the word on its own, eg ‘masquerade’ for ‘pose’; this could logically be extended to other ‘forced’ compounds, such as ‘the raw’, but one would have to find another word with a definition along the lines of ‘(with the) a sore or sensitive place’, which could be tricky. The nouns ‘push’ and ‘sack’ are a pair that spring to mind as being acceptable.
I hope that makes some sense.
Thanks, Doc, for those very helpful explanations. I was experimenting with QXW and it included RAWS in the grid. I remembered your mentioning questionable plurals in automated grid-filling software ( Setters’ Corner for Azed 2,780) and that prompted my question. If I ever get around to setting my first puzzle I’ll certainly be avoiding those types of words – and the multi-word expression components.
That would certainly be my advice – even if the answer gets past the editor, it’s unlikely to go down well with solvers. If a grid-filling program inserts a word like ‘raws’, I delete it from the word list (as I do with other words which I think might be considered ‘inappropriate’ for any reason – there are still plenty of others left!).
Incidentally, the Chambers entry for the noun ‘raw’ is interesting. It reads “n (with the) a skinned, sore or sensitive place; the raw state; that which is raw.” The ‘Using the dictionary’ section of Chambers implies that a label such as this applies only to the meaning to which it is attached. This is how the entry has been interpreted when it was imported to the electronic version of the dictionary in WordWeb:
noun
1. (with the) a skinned, sore or sensitive place
2. The raw state
3. That which is raw
The one sense of the noun which the OED gives without the definite article is the first of those above!
Hi Doc, and thank you for the fabulous resource. In Anagram Indicators, I think (?) you list participles with reference to transitivity. For example, “change” can be transitive or intransitive, so you list both “changed” and “changing”. Whereas “modify” is generally transitive-only, so you list only “modified”. I think (?) this is broadly speaking why some verbs are listed only as past participle, others only as present participle, others as both. My question: does this imply that you consider transitive-sense ing-form parse grammar to be incorrect? For example: Modifying dog for deity (3)
Hope clear, and apologies if stupid question!
Hi Guy, and thank you for your kind words.
Your question makes sense and is by no means stupid. It is impossible to be prescriptive, and whether a specific construction is valid can only be judged by an analysis of the cryptic reading (following which there will still be plenty of scope for subjectivity!). However, the active form of a transitive-only verb requires an object, hence ‘Dog modifying deity’ is not valid, while ‘Dog changing deity’ is fine. In your ‘Modifying dog for deity’ example, the fodder (DOG) is the object of the verb (good) but there is no subject (bad) – ‘Modify dog for deity’ (imperative to solver) or ‘Dog modified for deity’ (past participle) should be preferred. A construction which I see from time to time is exemplified by ‘Deity modifying dog’ (or ‘Deity modifies dog’); this seems superficially appealing, the answer being the subject of the verb, but ‘x modifying/modifies y’ is not the same as ‘x results from modifying y’. My advice would be to avoid active forms of intransitive-only verbs as indicators.
Note that in the (valid) ‘Dog needs modifying for deity’, the ‘modifying’ is a gerund (verbal noun) rather than a present participle, so the clue is equivalent to ‘Dog needs to be modified for deity’.
I hope that is helpful.
Many thanks for detailed reply! Understood that Azed-style strict Ximenean doesn’t allow this, helpful to know, and I get the reasoning. Meanwhile, from trawling a clue database, it seems to be quite common in weekday Times/Guardian etc (i.e. wordplay grammar is subject-less transitive-sense gerund phrase). I think you are saying that wordplay grammar must be a complete verb phrase, or alternatively a noun phrase. It might (?) be argued that “modifying dog” can potentially be a noun phrase, as in “[modifying the dog] was anti-ethical”. Anyway, not arguing against the Azed-type position, just trying to understand it and work out my own view. Thank you, really appreciated!
Hi Guy
Would you be able to provide a couple of examples from Times/Telegraph/FT puzzles of the sort of clue that you are talking about? I did a search for ‘modifying’ in George Ho’s database but (as I would have expected) only ‘after modifying’ and ‘needs modifying’ came up, while a search for ‘modified’ produced 37 clues where the word on its own indicated an anagram.
Hi
With homophone indicators, can the fodder come after the indicator? E.g. “Said when….. ” to give WEN
Hi Codujama
With indicators in general – and homophone indicators are no exception – it is accepted that adjectives, participles and adverbs can come either before or after the fodder, even when this appears to go against normal English usage. So the adjective ‘wild’ would normally precede the noun which it qualifies, as in ‘Wild West’ for STEW, but ‘run wild’ for URN would also be allowed. Similarly, for instance, ‘reportedly’ can go either side of its target.
Hence ‘said lines’ for ROSE [‘rows’] would be ok, as would ‘lines said’, and ‘broadcast live’ or ‘live broadcast’ would work for BEE [‘be’]. Any homophone indicators which must go on a particular side will fail a basic sense test when wrongly positioned, so ‘sound of hooves’ is fine for FEAT [‘feet’] but ‘hooves sound of’ makes no sense grammatically, and would need to be something like ‘hooves, by the sound of it’.
The main thing to remember with a homophone clue where the homophone and the answer are the same length is that it must be clear for an ‘either side’ homophone indicator which word (or words) it applies to. ‘Price for trip reportedly reasonable’ could lead either to FARE or FAIR, and is thus an unfair clue; for FARE it would need to read ‘Reportedly reasonable price for trip’, while for FAIR it would be something like ‘Reportedly, price for trip is reasonable’.
I hope that makes sense.
Many thanks for your time in producing that detailed reply. Very useful as usual
HI Doc,
Thank you for this superb resource.
I’m wondering if a homophone and wordplay would pass without a definition. For example, “Reportedly remained sober” should be fine for STAID, but would “Reportedly remained when good person came to help” also be acceptable?
Hi Bill, and thank you – very kind.
The one-word answer to your question is ‘no’. A clue must have at least one definition, in the sense of a word or words which lead to the answer without the use of any cryptic devices such as homophones, anagrams, reversals etc. A clue can legitimately have multiple definitions, or multiple wordplays with a single definition, so ‘Reportedly remained sober – it’s a wasted day’ would be ok for STAID, although you’re unlikely to see a clue like that in a UK back-pager.
Thank you, Doc
Hi.
I would be grateful if I could have some feedback on a clue I’m writing please.
Base alliance wants to entirely ruin you and do away with you it’s said? (5)
Please let me know if you want me to post my intended parsing here
Many thanks
Hi Codjuma
No need for parsing, but can you tell me the intended answer?
Hi. Thanks for responding. The intended answer is ALLOY
Just realised that “alliance” doesn’t work as the definition as it includes the majority of the answer letters. Also, wordplay wasn’t clean so I hope this re write improves those issues.
Base partnership entirely ruins you and does away with you it’s said? (5)
It’s important with any clue to give the same level of grammatical scrutiny to the cryptic reading as to the surface reading. Because we instinctively know what constitutes a sound piece of ‘regular’ English, in practice this means that the real challenge is the cryptic reading, which is more akin to algebraic notation.
In your original clue, the word ‘to’ had no place. The new version is a significant improvement, and having no link words between definition and wordplay makes achieving grammatical accuracy easier, as the wordplay can be considered on its own, ie ‘entirely ruins you and does away with you it’s said?”. ‘Entirely’ is fine for ALL, and the (anagram of YOU – homophone of ‘you’) for OY is potentially valid, but there is a problem with tenses: ‘entirely ruins you’ cannot be telling the solver to start with ALL and then rearrange YOU. The anagram indicator needs to be either an adjective/past participle (eg ‘ruined’) or the present participle of an intransitive verb (eg ‘breaking’). That can get us to ‘entirely ruined you’ for ALLOYU, but ‘[ALL + OYU] and does away with you’ isn’t grammatically sound. Since ‘ruined you’ is a noun expression, the answer is to further modify that expression, eg ‘OYU, doing away with you it’s said’. I don’t think there’s a need for the question mark on the wordplay, so a grammatically valid version might be “Base partnership entirely ruined you, doing away with you it’s said”. I hope that’s helpful.
Very helpful, thank you
Hi Doc,
Just toying with writing composite anagram &lits.
I came across the following (from a back pager…) which I doubt you would consider valid due to the lack of potentiality indicator:
1) This gin gets legions drunk / SLOE
If tweaked to the following, would you consider it sound?
2) This gin could get legions drunk (4)
I’m asking because all of the examples I’ve examined (mostly Azed of course) place the anagrind at the other end (as in “when drunk this gin could get legions” which obviously wouldn’t work for this clue surface).
In clue 2 above I feel I can read it cryptically as
solution + additional fodder could make [another anagram]*
…which strikes me as acceptable, but I’m not very familiar with these clues!
Thanks!
Hi Anon
Given the relative rarity of composite anagrams (&lit or otherwise), I think it’s fair to say that the rules are less clearly defined than for other clue types. It seems to me irrelevant whether the wordplay tells us that the answer and string A can be rearranged to form string B or vice versa – it comes to exactly the same thing. When blogging Azed puzzles, I would always pick the comp. anag. clues out for comment, because I know that they invariably cause difficulties for solvers (I’m not sure there was any clue of this type which didn’t prompt a question on one of the help forums) – part of the problem is that the definition is rarely at one end or other of the clue. I therefore believe that the solver needs to be given a bit of help so that they will (we hope!) enjoy the clue rather than being frustrated by it, hence my insistence on some suggestion of potentiality. This is often helpful when it comes to the &lit variety, such as the one in your example (I’m not aware of sloe gin actually making legions drunk), and one of the earliest examples, AF Ritchie’s “You could make this whale seem quarrelsome, but hold it up by its tail and it begins to laugh” for RORQUAL (the last 12 words are gratuitous by today’s standards). I think that something like Dr Fletcher’s “You might see this used in mixed doubles” for LOB is extremely fair (and the word ‘might’ keeps me happy).
Coming back to “This gin could get legions drunk”, I think it works very nicely as a composite anagram. As a clue, it would probably be acceptable to some editors of puzzles where composite anagrams are allowed. I say ‘some’ because of the definition – SLOE is not a gin, it is the distinguishing ingredient of SLOE GIN, in rather the same way that FRENCH is not a dressing; this would see it rejected by (say) the Listener editors. It’s the sort of situation where one solution is to replace the ‘This’ by a long dash, as in Mr Palmer’s “—— brood could be tutored by dons” for STUDENTY. Another possibility is to put the word ‘This’ is italics, with the same objective of telling the solver that it must be replaced by the answer prior to the clue being read as a definition of the answer. Either technique would, I think, work here.
In general terms, while I am probably rather less keen on the composite anagram than Azed, I would echo his observation that “the composite anagram, well handled, can be very neat, provided (a) the wording indicates clearly and accurately how the solution is to be arrived at, and (b) the extra letters to be discarded are not too numerous, swamping the key residue as it were.” When it comes to (a), the question a setter must consider is “If I came across this clue in a puzzle, would I feel that the signposting was sufficiently clear?”
Many thanks Doc, an incredibly helpful answer, as ever!
Hi Doc
What is your view on using the acronyms A and D for “across” and “down” respectively ? Any solver of crosswords ought to relate to that and you yourself refer to clues in grids in the same manner
Hi Avtaar
Fair question! While Across/A and Down/D would be unfamiliar to people who don’t solve crosswords, that is irrelevant given that we are talking about something that would only be seen in crossword clues. My first problem, though, is that the online Collins, with its remarkable array of abbreviations (A = anode/assist/attendance/arterial/affinity) doesn’t give them. Also, and perhaps more importantly, while we often use the abbreviations as part of the shorthand for discussing clues, they are rarely encountered in the puzzles themselves, and I have never (for example) seen the groups of clues headed ‘A’ and ‘D’. I don’t think that we can make an exception to the normal rules simply because the abbreviations belong to the world of crosswords, so I would not allow them in isolation, although the additional context provided by, say, ‘one across’ for IA might have me tossing a mental coin.
Hi. I distinguish between the abbreviations in Collins Online that are given as British English and those given as American English, which seem to be taken from Webster’s. I consider the American abbreviations as not valid for UK puzzles, and I had been presuming you did too, but your examples are from the American list, so maybe that is not the case?
Hi Andy
The point I was trying to make in this particular instance was that even among the vast array of abbreviations under various headings in the online Collins, many of which would not be allowed in any UK crossword, A for ‘across’ and D for ‘down’ are still conspicuously absent.
When it comes to UK crosswords, the abbreviations given by Collins are largely irrelevant (and those shown as ‘American English’ completely so). For barred puzzles, anything in Chambers is ok, and anything not in Chambers is not (with a few possible exceptions such as ‘small’ = S, ‘large’ = L, and ‘electric vehicle’ = EV depending on the puzzle series). For blocked puzzles, the UK newspapers typically have their own lists of allowable abbreviations, which are subsets of those in Chambers supplemented by a few extras such as ‘large’ and ‘small’. Something like ‘affinity’ for A (which Collins shows as a ‘British English’ abbreviation) would never be allowed, and nor would those ‘American English’ ones such as ‘attendance’ for A, which would seriously discomfit most UK solvers.
The abbreviations given by the ODE probably come closest to reflecting current usage, with (for instance) D for ‘drawn’ and L for ‘lost’, ubiquitous in the sports pages but not (yet?) allowed in crosswords.
Would you accept blocks/blocking as an insertion Ind ?
Yes, I would. Coincidentally, ‘blocks’, ‘blocking’ and ‘blocked by’ are on the waiting list of indicators to be added at the next update (probably next month, though there might be a delay as a result of the other changes that we are making to the site). I think their omission was simply an oversight.
A very commonly used indicator for first letter, basically, is not listed in Clue Clinic. Do you think it is valid in the sense of conveying that the first letter is to be taken? How about mainly and chiefly for initial letters?
Hi Kabir
That’s a good question, and it raises a point that I haven’t directly covered before. A search of George Ho’s database reveals that ‘basically’ is indeed often seen as a first letter selection (FLS) indicator, but only in the crossword in The Hindu paper; I couldn’t find any instances of it in UK cryptics. On the other hand, ‘fundamentally’ (another one which is not in the list on this site) has appeared occasionally in both Indian and UK puzzles, while ‘firstly’ (another absentee) is a regular in UK cryptics (which makes me think that it should be added to the list).
Adverbial indicators in general are by their nature somewhat questionable, not just because they cannot be applied directly to a noun (so a verb inflection like ‘selected’ must be inferred) but also because their true meanings ‘in real life’ are not what is required in a cryptic context. One could certainly argue that there is no significant difference between ‘initially’, ‘originally’, ‘primarily’, ‘basically’, ‘fundamentally’ and ‘firstly’, and that if one is valid for FLS then they all should be. However, precedent – and thus solver expectation – is important when there is a major divergence from real-life usage, and for that reason I would suggest that ‘basically’ would cause no problems to a Hindu Times solver, but would probably not go down well with a Daily Telegraph solver. In any event, though, I would expect the editor of a given puzzle to change indicators such as these which they believe will not be appreciated by the solving community.
Similarly, I think that precedent/expectation is key when it comes to ‘chiefly’ and ‘mainly’. They can both mean ‘in the first instance’ and ‘for the most part’, but on the rare occasions when ‘chiefly’ occurs it is as an FLS indicator, while ‘mainly’ is seen frequently, but as a last letter deletion indicator (similar to ‘largely’). I would therefore definitely avoid ‘mainly’ for FLS purposes (very likely to confuse solvers), and would use ‘chiefly’ with caution – unless you know that your solvers are comfortable with it.
Thank you
You’ve reminded me that ‘extremely’ for first and last letters is ubiquitous in British cryptics, but it wasn’t till I found out that Azed didn’t use it, and wouldn’t accept it in a competition clue, that I realised that he had a point.
Anyway, on more or less the same subject, do you have an objection to ‘eventually’ as a last letter indicator? I guess it’s not generally accepted – I’ve seen it only once, years ago.
And why are ‘across’ and ‘over’ not in your containment list? I’m guessing that’s deliberate as they’re similar.
Hi CG
A rational analysis might conclude that ‘eventually’, like ‘during’ as a potential insertion indicator, is only ever used in a temporal sense; also that ‘final’, ‘last’ and ‘ultimate’, unlike ‘eventual’, can all function as nouns, so eg ‘last of X’ and ‘X lastly’ could (in cryptic-speak) perhaps be seen as broadly equivalent. In reality, though, my biggest objection is that – as you say – ‘eventually’ is simply never seen in that role.
I view ‘over’ and ‘across’ as two prime examples of indicators whose use is justified by interpreting a bald Chambers definition in a way that suits the setter’s purposes, specifically ‘from side to side (of)’ for ‘across’ and ‘from side to side or end to end of’ for ‘over’. But neither ‘across the sea’ nor ‘over the sea’ suggests that the sea is being contained, and I’ve never found anything in the OED which supports their use as containment indicators. They are both in the Lexicon, marked as being of questionable validity.
Thank you, and sorry I hadn’t noticed the empurpled across and over in the Lexicon section. (But I’m grateful for the opportunity to use ’empurpled’.)
I remember Sunday Times editor Peter Biddlecombe justifying ‘over’ as a container by likening it to a bridge going over a river. Whether that would mean ‘under’ could be an insertion indicator [cont. P.94]
😀 I hope that following the site revamp we’ll be able to come up with an improvement on the ’empurpling’!
The bridge going over the river invokes a different Chambers definition, ‘above and from side of to the other’, which doesn’t suggest containment to me. It’s a grey area, undoubtedly, but I do feel that prepositions should be subject to particular scrutiny, given the number of unexemplified definitions often given for them by Chambers – eg 17 for ‘over’, 24 for ‘of’ including ‘from’, ‘with’, ‘over’, ‘by’, ‘on’ and ‘in’.
“Over” used to signify containment in a down clue is a pet hate of mine. Horribly misleading I think!
Hi Doc and Kabir,
Thanks for raising this – I’m intrigued by the argument for “basically” as I have seen others use it in competitions. (Now I know why!)
I realise these things are about conventions, but I can’t really see how “basically selected” would *unambiguously* signal selecting the start or first of something. The basis / base of something (etymologically speaking) refers to its foot/bottom, no? Of course, many things develop from the bottom up (like plants) and figuratively this lends a sense of “beginnings” to base, but for me there’s far too much conflict between both meanings (start vs foot) to make it viable.
“Fundamentally” carries a stronger sense of origin/starting although I think it’s still a stretch, because again it relates to words like foundation – i.e. base…
Chiefly does have roots in Latin caput, meaning head, which I think is a solid argument for it to be used as an FLS indicator (rather than a last letter deletion indicator, which has always struck me as tenuous).
The problem, for me, with looking for justification from Chambers is that it often uses synonyms to define words, and while words like “basically”, “fundamentally” and “principally” may be interchangeable in many situations, they do have subtly different nuances that reflect their etymology.
On the other hand, “foremost” is listed in most dictionaries (sadly not Chambers) as an adverb. I think it could reasonably be used as an FLS indicator, given its derivation. (But of course it isn’t, thanks to Chambers…)
In all honesty, I find it easier to make a case against, rather than for, pretty much every adverbial FLS indicator, which is why I am largely prepared to be guided by custom and practice. It did, though, strike me as I was writing my reply to Kabir that ‘firstly’ not only has claims which are at least as good as most of the other similar indicators in the list on this site but also – unlike, say, ‘basically’, about which I share your doubts – has just one possible meaning as an indicator, such that even if it can’t be technically justified it’s unlikely to be misunderstood.
Yes I can see that. Curious as to why “firstly” was omitted from the list. Is it just because “barred setters” (for want of a better term!) tend to avoid it?
Every now and then a (relatively) glaring omission from the indicator lists crops up. The early lists were the result of a brain dump, followed by a check of a few established reference sources, followed by a trawl of a considerable number of Azed Slips, from which I selected only those indicators which I felt at the time were valid (a few that got through my filter have subsequently been weeded out). I see that there is only one published Azed competition clue that uses ‘firstly’, so I’m quite likely to have missed it. I think that ‘initially’ tends to be the FLS adverb of choice in clue-writing comps, and generally sound more natural than ‘firstly’; it’s a bit different in complete crosswords, because you may already have exhausted your preferred options in other clues.