Blog post by Scott Mainwaring
The flag of Provo, Utah ranked extremely low (143rd out of 150) in the flag designs evaluated in NAVA’s 2004 American City Flags Survey, eventually leading to an effort in 2013 by Mayor John Curtis to create and adopt a new flag for Utah’s third-largest city.

The effort got bogged down in a long, drawn-out process. Reporting on a Municipal Council work session in September 2014, the Provo Daily Herald noted that even after a “full year-and-a-half […], many discussions, resident comments, blogs and more vetting… the municipal council has yet to figure out just what long-term symbol could represent Provo on the city’s flagpole.” Indeed, within the three-person committee tasked with recommending a design, a dispute had arisen between “Provo’s resident vexillologist, Jason Bates” and graphic designer, city councilman, and designer of the flag to be replaced Stephen Hales. (The position of the third committee member, former councilwoman Sherrie Hall Everett, isn’t mentioned.)
Jason Bates cited the Good Flag, Bad Flag principles and the flags of Chicago and of the District of Columbia as particularly good exemplars. The Daily Herald quotes him as saying:
It’s important to come up with a good flag design. A good city flag design can become more than just a logo.
Stephen Hales, on the other hand, didn’t “seem to be as caught up in flag tradition as his team member Bates,” saying:
Have the flag with a circle in the middle with the city’s logo. It doesn’t need classic flag elements. It needs something that extends the city’s identity.
In an effort to move forward the mayor’s office offered two designs to be considered, each referencing elements from the city’s logo but not including it in its entirety:


The work session adjourned without reaching a decision. By the next month, the committee had made a decision. Current and former council members Hales and Everett outvoted vexillologist Bates, 2 against 1, and Hales presented a design based on his own suggestion from the previous session (having passed over both of the mayor’s office proposals). Hales’ new design went on to be approved on January 6, 2015 as Provo’s new flag, replacing Hales’ old design:

What happened to Bates’ and the mayor’s office proposals? Anyone interested in the politics of municipal flag design will find this video record of the Provo Municipal Council work session of October 21, 2014 to be most informative:
What happened at the work session provides a number of lessons for flag designers and vexillonaires:
(1) Hales asserted that his design was consistent with NAVA’s professed principles of good flag design by referencing highly-rated municipal flags in the 2004 American City Flag Survey that included circular logos (like Phoenix, ranked 4th) or what he interpreted as circular logos (like St Louis, ranked 5th).


(2) Flag design principles themselves were deemed more or less irrelevant. We are told that they might have been historically important for military flags, but what has this got to do with modern day municipal flags like Provo’s? Such principles might be important for a flag lasting 50 or 100 years, but this is outweighed by the needs of the immediate present; it’s much more important, we are told, that the flag be immediately recognizable without any effort by current residents. They might be important for flags that are embraced by a population and flown by many private citizens; but the flag to be designed has no such lofty ambitions — if it’s only flown from flag poles the city owns, it will be a complete success. (In other words, the flag is not for the people of Provo, but for the government of Provo — it’s a service provided to them, nothing more.)
(3) Related to the above, it is clear from this video that the flag was seen as properly subordinate to a larger effort to brand the city: a larger, expensive, professionally designed marketing effort. Seen in this light, it makes perfect sense to defer to the judgement of an advertising professional (not an amateur vexillologist) and to put a great deal of weight in evaluating a design the degree to which it leverages the city’s substantial investment in a logo (and not the degree to which it draws upon more enduring symbols than logos).
(4) The discipline of flag design isn’t doing itself any favors, in these political contexts, by branding itself with arcane, hard-to-pronounce, and apparently hard to spell terms like vexillology.
(5) Good design by committee, if not impossible, is very hard! In the case of Provo, as in many others (e.g., see the case of Toronto in Vexilloid Tabloid #47), it often comes down to a single person vigorously promoting their own design.
At the end of the day and after tiring effort and controversy, Provo has a new flag, one that is almost certainly better than the one it is replacing. It’s possible that the people of Provo will come to love this logo-based flag, and to adopt it as their own, not just as city hall’s. It’s a shame, however, that this populist criteria for defining what makes an enduring, embraceable “good flag” was unable to carry the day against more marketing-based, short-term competitors.
Not bad, not so good either. No alphanumerics. Could have been worse. When it comes to flags, it can get personal. Patrick G.
Hi
[…] year Republic is considering updating its flag. Unfortunately, as in the case of Provo, Utah, the main concern seems to be leveraging the city’s investment in a new logo (a stylized R) […]
[…] Apparently Fargo’s government is making the mistake of seeing the city flag only as their flag, rather than the people of Fargo’s flag — and thus requiring coordination with the city’s professional branding efforts, and likely greatly limiting its popular adoption (see also: the case of Provo). […]