Cfxtrjtrk

Both spellings are used depending on the variety of English. According to Wiktionary:

The spelling focused is much more common in the US; however, the spelling focussed is sometimes used in the UK and Canada, and is especially common in Australia and New Zealand.

According to the website of a UK-based company Future Perfect, the general rule is as follows:

The official requirements are that we ‘double a single consonant letter at the end of any base where the preceding vowel is spelled with a single letter and stressed’.

(Here, if the preceding vowel is the only vowel in the word, it is counted as stressed.)

However, I do not know how the spelling “focussed” fits this rule. Maybe it is an exception to the rule.

I believe the rule is that you double the final consonant when both of the following are true:

• the consonant ends a stressed syllable or a one-syllable word


• the consonant is preceded by a single vowel

As 'cus' is not the stressed syllable, it would not be doubled according to this rule.

I believe the 'stressed' requirement is relaxed in some situations under British English, however. Indeed, I found references to 'focussed' with an 'especially British' tag.

I've actually thought a good bit about this over the years, and I find it interesting that not a single site, including very authoritative ones from British and American lexicography or journalism sites, really fully describe and explain this properly. Naturally, this is my own interpretation, but I think the rule covers it, with very few exceptions, and even those can be explained via the normal tendency of native speakers to unconsciously parse or analyze such situations and come up with a solution based on prior experience and examples as to what feels "right"; this is a normal part of language development, and can explain what are otherwise seen as exceptions. Of course, as a descriptivist, such usage will become, or is, the rule, irrespective of how you try to formulate a "rule" to cover all cases. But, with those caveats, this is how I see it:

American rules

Double final consonant on verbs which:

1. end in consonant + vowel + consonant (excluding final -w or -x) (n1)


2. and have regular past tense formation (n2)


3. and have the stress or accent either:
   
   
   
   
   • on the last syllable in a simple or a compound verb (n3)
   
   
   • or on a prior syllable in a compound verb in which the final syllable:
     
     
     
     
     • is the root verb (n4)
     
     
     • is not the root verb, but is falsely analyzed as if it were (n5)
     
     
     
     
   
   
   
   


4. With some exceptions (that may not really be exceptions) (n6)

Notes

1) Typical examples:

• overdub embed defog repel retrim outgun drop reship bar defer commit emit bus gas regret rev;


• but not: catalog travel abandon button darken happen worship caucus focus edit profit.

But also probably including some rare final consonants as part of the regular pattern: -c -f -h -j -k -q -z

"He sicced his dog on me." is acceptable, "He siced..." is not. Although another rule says to add -k to final -c to make -ck, that's only the case for unstressed final syllable, as in, mimicked, frolicked, panicked. So, not "He sicked his dog..."

Note that a way of "testing" doubling of rare final consonants, is with the "out-" trick. In English, a verb can be constructed by using "out" + any person's name (or title). Example, using a non-rare final -d: "She out-Crawforded Joan Crawford." Now test -h: which looks right?

• The new ruler out-Shahed the Shah.


• The new ruler out-Shahhed the Shah.

Not clear to me, but I think I prefer the second, which would put -h into the regular group that gets doubled when stressed; what do you think?
When I try the "out-trick" with final -c -f -h -j -k -q -z, none of them seem like exceptions to me. In the end, these rare cases don't really matter because they don't come up in the real world, and no rule can apply to single or constructed examples, although it's interesting that my "native competence" has me going with regular formation for all of these: "out-Jean-Luc'ed" (not the final syllable!), out-Krystofed/out-Steffed, out-Elijahed/out-Shahhed, out-Andrej-ed/out-Rajjed, out-Dereked/out-Zakked, out-Tawfiqed/out-Shaqqed Shaquille, out-Hafezed the poet, but out-Bozzed Charles Dickens.

2) Rut, but not cut; rebid, but not forbid; regret, but not forget.

3) Refit, but not profit. One-syllable words are accented by definition on the "last" syllable, so are included.

4) ramrodded, deadpanned, suntanned, giftwrapped, checksummed.

5) false or uncertain compound analysis:

• eavesdrop: eavesdropped, probably assuming "eaves" + "drop" but that is false as eavesdrop is a backformation from eavesdropper.


• kidnap: kidnapped, probably also a backformation from kidnapper, but "nap" (snatch) is a verb and has the slang meaning underlying "kidnap" so the compulsion to use --pped here is strong


• worship: worshipped, not from the one syllable "ship, shipped" but from -ship, a suffix which cannot stand alone and derives from a Middle English suffix: apprenticeship, citizenship, courtship, guardianship, kinship, lordship all use this same suffix.

6) Exceptions:

• programmed: looks like an exception, but is it? Some say it's about a phonetically-based orthographic change, so it doesn't look like it rhymes with "gamed", but I think it's because the English word is "programme", and so the past is just adding a final -d, and so this word isn't part of the rule at all, because doesn't comply with condition one. Also, "monogrammed' and other -gram/-gramme examples. 'Program' predates 'programme' I believe, but I think verbification succeeds it, especially in the computer sense which didn't happen till mid-20th century.


• there are some others, which I forget right now, but I'll add if I come up with them later.

British English

Same as American English, as far as conditions 1–3, except:

• Verbs ending in -l which are not accented on the last syllable, still double the final consonant. So:
  
  
  
  
  • travelled, cancelled, modelled, labelled, and so on.
  
  
  • As an AE user primarily, these seems strange to me, especially the last one; when I look at that one, I can't help think of the singer. (She out-labelled Patti..."). But then, I'm sure BE users think our way is weird.
  
  
  
  


• there are a few exceptions where you don't double unstressed final -l in BE, but I forget what they are.


• I've seen some sites claim that final -p is a kind of BE difference, but I don't think so, because the usage is pretty similar to AE, and I think both are subject to the misanalysis in 3 (b) (ii) above, and so, BE usage is not really different from AE usage here, and neither is it really an exception.


• I've seen final doubled -t, as in ''profitted'' or ''benefitted''; not sure if this is BE, or just a mistake.

Would love to hear feedback or comments on this from y'all. BBC, Reuters, M-W, A-H, or Oxford: if you pick this up, I want a credit! (Bryan G/Noad: you can have it for free; you've helped me enormously over the years, my turn to return the favor.) ;-)

Here is how I would formulate the rules for "consonant doubling" in English words or proper nouns that have had a vowel-initial suffix attached. (The most common suffixes that regularly cause doubling are -ing, -ed, -er, -able, -est, -y; less frequent suffixes that can cause doubling are -ist and -ish.)

1. 
   

   The following letters are never doubled by this rule: a e i o u w y.

   
   
   

   x is normally never doubled, but it might be possible with some neologisms or informal words like "exxed" "crossed out with an X".

   
   
   

   h is less obvious, since so few words end in vowel + h, but I think that it also is not included in this doubling rule. The only "-hhed" spelling I found that seems at all common is "ahhed", where the base could be interpreted as "ahh". "Blah-blahed" and "hurrahed" seem to be spelled without doubling.

   
   
   

   Other letters are typically subject to the doubling rule. The rule is less clearly established for letters that are rare or extremely rare in word-final position (v j q), but we do see doubled spellings in e.g. shivved, so v at least seems to be subject to the rule.

   
   



2. A letter must be word-final in the base word to be doubled: there are no words that remove a final "silent e" and also double the preceding consonant letter. So we never double the consonant in words like giving, promising, disciplining.*



3. 
   

   Doubling is only possible when the consonant letter comes directly after a single vowel letter that by itself represents the vowel of the last syllable. This means that:

   
   
   
   
   
   • Doubling does not occur after vowel digraphs like "ea" or "ai", regardless of how they are pronounced.**
   
   
   • Doubling may occur after a single vowel letter preceded by "qu" (quipped") or by another vowel in hiatus ("duetted").**
   
   
   • Doubling does not occur in words that end in more than one consonant letter, whether that is a digraph (slash, graph, path) or consonant cluster (twist, act).
   
   
   
   


4. 
   

   Doubling usually doesn't occur when the final syllable is unstressed, but there are various exceptions, mostly but not exclusively involving words that can be analyzed as having a "minor" stress on the last syllable (many of these are compounds or prefixed verbs). A number of the exceptions may also be written without doubling.

   
   
   

   In British English, there is one systematic exception to this stress-based rule: double ll is used even after an unstressed vowel. But the other rules that I mentioned in 2 and 3 still apply.

   
   
   

   Focussing is another exception to the stress-based rule; writing it with a double consonant is optional.

   
   

* If you include plural nouns, then "premisses" might be a possible exception to this, but doubling is so rare in plural nouns that I don't think we can say that it follows the same rule. (I wrote a separate answer about doubling in plural nouns: Why is the plural of “quiz” spelled with double Z?) "Premisses" is also a special, complicated case.

** Sometimes it may be tricky to tell apart vowel digraphs and vowels in hiatus. British spelling uses single l in words like "healing", "stealing", "mailing". But because words like "gruel" and "fuel" are traditionally pronounced with two syllables, the spellings "gruelling" and "fuelling" are used in British English. For the same reason, I think the adjective "real" should theoretically form "realler" and "reallest" according to the British rule, but single-l spellings of these forms seem to be currently more common even for British writers.

The double 's' is an aberration, an abomination and makes my eyes bleed when I read it - I posed the same question to the Oxford Word and Language Service ('OWLS') in 2009, and they replied, quoting chapter and verse (as previous contributors have), that the correct usage is single 's' but that 'some' British printing styles allow for the double 's'. Hence, it is accepted by word processing spell checkers, and people think their preference is 'right' because the spell checker lets them get away with it. Thus begin habits and opinions, uninformed by grammatical rules. Many think it is yet another difference between American English and British English, and reason that if American is single 's', British must surely be the other version - as you'll no doubt know, it is the same in both - single 's' rules!

For reference, the Guardian & Observer style guide simply states that the correct usage is single 's' (http://www.theguardian.com/styleguide/f), whilst the Economist style guide does not get around to it, stating that the word 'focus' is overused.
- So stay single out there!