From Arnold Zwicky of Stanford University:

For at least 35 years, English speakers have been producing sentences with an occurrence of a form of BE that is not licensed in standard English (SE) and is not a disfluency – our Extris (“extra is”).  There are many subtypes, but we observe that all are based on SE constructions with a specific discourse function and suggest that any SE construction with this function can have an Extris counterpart.   The Isis (“is is”, “double is’, etc.) subtype has gotten much attention – from Bolinger through Coppock et al. – as a variant of SE “thingy”-N-subject or pseudocleft (PC) sentences:

(1)  N-type Isis: The funny thing is is that Lisa was there too.

(2)  PC-type Isis: What’s nice is is that it has a sort of other-worldly character.
There are also Singlis (single-“is”) examples, where the SE counterparts are not copular.  In one
set (Jehn, Ross-Hagebaum), the clauses are deictic or existential – our Th:

(3)  N-type Th: That’s/Here’s our suggestion for it is that...

(4)  PC-type Th: That’s/This is what we hear all the time is that...

(5)  There Th: There’s one thing I need to do is leave a check.
Then there are McConvell’s (2004) FreeBe’s, in which initial material is either explicitly
cataphoric (as in (1)-(5)) or implicitly so:

(6)  Exp FreeBe: We looked at it this way is that...

(7)  Imp FreeBe: I’d like to say is that...
What unites (1)-(7) is that they are all variants of SE constructions that introduce content
by announcing, in an explicitly or implicitly cataphoric expression (SU: “set-up”), that it is about
to be introduced, and then supplying it in a following expression (PO: “pay-off”,  a.k.a.
“counterweight”).  The SUs are variously phrasal, hypotactic, and paratactic:

(8)  Simplex: The problem is (that) it’s time to leave.  (1)

(9)  Pseudocleft: What I think is that it’s time to leave.  (2)

(10)  Paratactic Apposition:
    (a) That’s/Here’s the problem: it’s time to leave.  (3)
    (b) There’s one thing I need to do: leave right now.  (5)
    (c) I’m telling you: it’s time to leave.  (7)

What the extra form of BE does in the Extris examples is explicitly mark the PO part of the SU+PO construction and so focus on it.  In any case, it seems likely that every sort of SU+PO construction (including some not listed above) has an Extris counterpart for at least a few speakers. In the other direction, extraneous forms of BE don’t seem to occur, except as disfluencies, anywhere but in SU+PO constructions.  You don’t find things like (11) Reading Sherlockian pastiches is is what I do to relax.  Extris versions are potentially available for all SU+PO constructions, but speakers differ as to which ones they use.  Many have none.  Some have fairly high rates of Isis, but no Singlis, and some have moderate rates of Singlis (of certain sub-types), but no Isis.  And there’s at least one who seems to be a near-categorical user of Extris, of all types.  The Extris types have a common function, but they are independent constructions.



Net discussions of Isis:

Mark Liberman,  27 June 2004. The thing is is people talk this way...
   http://itre.cis.upenn.edu/~myl/languagelog/archives/001123.html

Mark Liberman, 29 June 2004.  Isis Fest, with emergent free-bees
    http://itre.cis.upenn.edu/~myl/languagelog/archives/001135.html

Adam Albright, 29 June 2004.  A bird in the hand is, is...
    http://itre.cis.upenn.edu/~myl/languagelog/archives/001139.html

Arnold Zwicky, 5 July 2004.  Isis bibliography
    http://itre.cis.upenn.edu/~myl/languagelog/archives/001164.html



References

Andersen, Gisle. 2002. Corpora and the double copula. In Leiv Egil Breivik & Angela
Hasselgren (eds.), From the COLT’s mouth... and others’: Language corpora studies: In
honour of Anna-Brita Stenstrom (Amsterdam: Rodopi), 43-58.

Bolinger, Dwight L.  1987.  The remarkable double Is.  English Today 9.39-40.

Brenier, Jason M. & Laura A. Michaelis. 2005. Prosodic optimization via syntactic amalgam:
Syntax-prosody mismatch and copula doubling. Corpus Linguistics and Linguistic
Theory 1.45-88.

Coppock, Elizabeth; Jason Brenier; Laura Staum; & Laura Michaelis.  To appear.  ISIS: It’s not
disfluent, but how do we know that?  BLS 32.

Jehn, Richard Douglas.  1979.  That’s something that I wouldn’t want to account for, is a
sentence like this.  Calgary WPL 5.51-62.

Massam, Diane.  1999.  Thing is constructions: the thing is, is what’s the right analysis?  English
Language and Linguistics 3.2.335-52.

McConvell, Patrick.  1988.  To be or double be?  Current changes in the English copula.
Australian Journal of Linguistics 8.287-305.

McConvell, Patrick.  2004. Catastrophic change in current English: Emergent Double-be’s and
Free-be’s.  Talk at Australian National University.  Slides available at:
http://ldc.upenn.edu/myl/anubbppt3.pdf

Ross-Hagebaum, Sebastian.  2005. “And that’s my big area of interest in linguistics is discourse”
- The forms and functions of the English that’s X is Y-construction.  BLS 30.403-14.

Shapiro, Michael & Michael C. Haley. 2002. The reduplicative copula is is. American Speech
77.3.305-312.

Tuggy, David.  1996.  The thing is is that people talk that way.  In Eugene Casad (ed.), Cognitive
linguistics in the redwoods: The expansion of a new paradigm in linguistics (Berlin:
Mouton de Gruyter), 713-52.