This record is a partial extract of the original cable. The full text of the original cable is not available.

C O N F I D E N T I A L SECTION 01 OF 03 BAGHDAD 004200 
 
SIPDIS 
 
E.O. 12958: DECL: 10/11/2015 
TAGS: PGOV PNAT PHUM IZ
SUBJECT: WATCH SALAH AD-DIN AND NINEWA PROVINCES TO 
DETERMINE FATE OF THE DRAFT CONSTITUTION 
 
REF: A. (A) BAGHDAD 3898 
     ¶B. (B) BAGHDAD 4090 
     ¶C. (C) BAGHDAD 4143 
     ¶D. (D) MOSUL 149 
     ¶E. (E) BAGHDAD 572 
     ¶F. (F) KIRKUK 205 
 
Classified By: Political Counselor Robert S. Ford for 
Reasons 1.4 (B) and (D). 
 
¶1.  (C) SUMMARY.  A last-minute political compromise 
to elicit Sunni Arab leader public expressions of 
support for the Constitution could help ensure its 
passage in the October 15 referendum.  That deal 
apparently is sealed.  Were it to fall through, the 
outcome of the referendum will probably be very close 
in a couple of predominantly Sunni Arab provinces.  If 
three provinces see 2/3 of voters reject the draft, it 
will fail as per the Transition Administrative Law. 
Poll results and other data indicate that three 
provinces - Anbar, Ninewa and Salah ad-Din - may vote 
against the draft by sufficient margins to reach that 
2/3 threshold.  Polling, done well before the last- 
minute constitution deal, suggests Anbar and Ninewa's 
large Sunni Arab majorities will vote solidly against 
the constitution.  Ninewa's Kurdish minority appears 
too small to stem the anti-constitution vote in Ninewa 
alone.  Salah ad-Din, a province with a Sunni Arab 
majority and a significant Shia minority, would appear 
to be the key swing province.   Violence that targets 
and intimidates Sunni "no" voters or Shia "yes" voters 
could be decisive in moving the results a couple of 
percentage in points in either direction.  If public 
backing for the constitution from Sunni Arab Islamist 
leaders at the last minute swings a few percentage 
points of the Sunni Arab vote, that too could be 
decisive in keeping Ninewa and Salah ad-Din provinces 
from reaching the 2/3 rejection threshold.  End 
Summary. 
 
Poll Results Consistent 
----------------------- 
 
¶2. (C) Three statistically valid polls conducted 
during September by the International Republican 
Institute (IRI), INR, and the UN consistently show 
that the no vote on the October 15 referendum is 
significant in the heavily Sunni Arab populated areas 
of Ninewa (ranging from 89 to 95 percent), Salah ad- 
Din (from 64 to 77 percent) and Diyala (36 to 49 
percent).  No polling was conducted in Anbar province 
for security reasons.  If two-thirds of voters in 
three or more provinces vote no in the referendum, the 
Constitution will be defeated, triggering elections 
for a new transitional government. 
 
Anbar 
----- 
 
¶3. (C) The only data available from the heavily Sunni 
Arab Anbar province, where security considerations 
prevent regular polling, comes from surveys conducted 
by the National Democratic Institute (NDI) of 
participants in constitutional outreach events in 
September and October.  Though non-scientific (the 
sample is not unbiased), it does provide a measure of 
the views of likely voters who have taken the time to 
become more acquainted with the provisions of the 
draft Constitution.  Over 91 percent of 2,551 survey 
respondents in Anbar planned to vote no, up from 84 
percent of 1,317 respondents a week earlier. 
Provincial and Baghdad contacts widely expect the 
constitution to fail there. 
 
Ninewa 
------ 
 
¶4. (C) The results of all polls from Ninewa province 
consistently show an overwhelming no vote among likely 
referendum voters.  The UN poll of very likely voters 
in Mosul puts the no vote at over 95 percent; an INR 
poll puts the no vote in the city at 92 percent. 
IRI polling in the entire province puts the "no" vote 
at 94 percent. INR puts the provincial "no" vote at 89 
percent.  Since there is a sizable Kurdish and 
Christian minority in the province, the scale of the 
"no" vote (ranging from 89-95 percent) raises 
questions about bias in the sample size.  The 
unscientific NDI survey results show a more modest 
(though still significant) no vote of 79 percent of 
5,962 respondents, down from 81 percent of 1,799 a 
week earlier. 
 
¶5. (C) Only 17 percent of eligible voters turned out 
for the January TNA elections that were generally 
boycotted by the Sunnis (compared to turnouts of 92 
percent and 84 percent respectively in the heavily 
Kurdish provinces of Dohuk and Erbil).  This suggests 
that more than 80 percent of the voters in Ninewa may 
be Sunni Arabs.  If most Ninewa Sunni Arabs turn out 
on October 15, and roughly 75 - 80 percent vote 
against the draft, the outcome will be near and could 
surpass the all-important 2/3 threshold.  Our 
Christian community contacts suggest the Christians 
will vote against the draft too; if they do so in 
large numbers they will help cancel out the Kurdish 
"yes" vote (reftels C and D). 
 
¶6. (C) There were credible charges of vote rigging and 
other electoral irregularities in Ninewa province in 
January.  The Independent Electoral Commission of Iraq 
(IECI) is using Iraqi security forces to help deliver 
materials - just as it did in January when these 
forces were accused of unscrupulous vote rigging (see 
ref E).  There is, therefore, the risk again of vote 
fraud.  Fuad Hussein, Chief of Staff to Kurdish 
President Barzani, confidently told PolCouns October 9 
that the pro-constitution camp will prevail in Ninewa. 
PolCouns strongly cautioned Hussein against vote 
irregularities such as those seen in January, urging 
Hussain to help ensure the referendum is credible. 
Hussein appeared unimpressed. 
 
Salah ad-Din 
------------ 
 
¶7. (C) Polls show Salah ad-Din on the brink of a "no" 
vote, which given the outlook in Anbar and Ninawa 
could spell defeat for the Constitution.  However, the 
margin in Salah ad-Din is close enough to make the 
province the key battleground in the referendum.  The 
UN poll of very likely voters in Tikrit in early 
September showed 64 voting no with 26 percent 
undecided.  The IRI poll from mid-September indicated 
that 66 percent would vote "no" in Salah ad-Din. 
INR's poll from the third week in September put the no 
vote at 77 percent for Salah ad-Din provinces and 49 
percent for Tikrit/Baquba.  There was a slight hint 
that the number of persons planning to vote against 
the constitution in Salah ad-Din was dropping in early 
October:  an unscientific NDI survey in October 
indicated that 66 percent of 6,008 respondents would 
vote against the draft, down from 74 percent of 3,983 
respondents a week earlier.  Only 29 percent of 
eligible voters took part in the January 2005 TNA vote 
indicating a sizable Sunni bloc in the province. 
 
Diyala 
------ 
 
¶8. (C) Poll results in Diyala, which show a no vote 
ranging from 36 to 49 percent among very likely 
voters, are still sufficiently high to warrant 
attention.  Only 33 percent of eligible voters turned 
out in the January TNA election indicating that a 
sizable Sunni electorate may be able to swing the 
district into the no category, though likely not with 
a sufficient margin to defeat the Constitution.  The 
IIP is campaigning hard in Diyala to mobilize a no 
vote (reftel F). 
 
Kurds Confident, Shia Apprehensive 
---------------------------------- 
 
¶9. (C) Kurdish leaders such as Fuad Hussein (para 6) 
are publicly confident that the referendum will be 
successful.   However, Shia and Kurdish leaders have 
expressed worry that carefully targeted terrorist 
attacks will deter Shia voters from getting to the 
polls (reftel B).  Notably, they highlighted concerns 
about Salah ad-Din.  In response, the TNA passed a 
since-reversed highly prejudicial resolution that 
defined voters in such a way as to ensure approval of 
the constitution.  The incident underscored the 
uncertainty, despite confident utterances, felt by the 
Kurd/Shia governing coalition about the prospects of 
voter intimidation in Salah ad-Din and other areas. 
 
¶10. (C) Well-connected Shia politician Ali Debbagh 
told PolCouns on October 10 that Grand Aytollah 
Sistani will issue a special statement to urge his 
followers to vote yes in the referendum.  The call has 
been delayed because, as one Shia interlocutor put it, 
"the Constitution is not yet finished." 
 
Tipping the Balance - Turnout, Violence, Fraud 
--------------------------------------------- - 
 
¶11. (C) Many Sunni leaders, especially from the 
National Dialogue Group and the Iraqi Islamic Party 
had gone public with their campaign urging voters to 
reject the constitution.  National advertising 
campaigns, including newspaper ads and posters urging 
a "no" vote are in full swing in many Sunni Arab 
areas. 
 
¶12. (C) The last minute political compromise to garner 
Sunni Arab politicians' public support for the 
Constitution could help chances of its passage in the 
referendum.  This would be particularly true if the 
head of the Sunni Waqf instructed imams to publicly 
express support for the compromise, something he has 
privately indicated he may do if the Iraqi Islamic 
Party also supports the draft.  From the above polling 
data, it appears that a swing of even a few percent in 
the Sunni Arab vote could keep Salah ad-Din and Ninewa 
from reaching the critical 2/3 "no" voter threshold. 
Barring that, other factors could tip the balance. 
Violence in the next few days in key Sunni cities such 
as Tikrit, Samarra, and Mosul could constrain the 
Sunni Arab voter turnout.  Similarly, violence against 
Shia - more likely - in Salah ad-Din towns such as 
Balad, Bayji, Tuz Khurmato and Dijayl, and Diyala 
towns like Balad Ruz, Baqubah, and Khalis could limit 
the Shia "yes" vote turnout. 
Khalilzad