Re: bias?


 

Hi Sean,

Yes, I can see how my statement was something of an invitation to do a numerical breakdown as you did. But why count species at all? What makes that the relevant taxonomical unit for assessing whether a group is overrepresented or underrepresented in awards? We could count botanical series, we could count the size of the areas where each type occur in nature, we make some kind of estimate of the number of individual plants alive at this moment; there are a boundless number of ways to assess "how much" of the genus any given kind represents.

Since the purpose is to assess whether awards are distributed fairly, it seems the only measure that is actually relevant is the number of cultivars available. The awards are given to cultivars, not to species. So we come back to my premise that the only sense I can make of "fair" in this context is that the opportunity for receiving an award is the same (at least approximately) for any new cultivar, regardless of its type.

I share your opinion that it's a shame that the potential of so many species as garden plants has gone untapped, and think it's a shame that so many gardeners think only of TBs when they think of irises. It's just that I think you have the cause-and-effect of it backwards. Awards can help nudge work in a particular class, but they cannot make any obscure corner of the genus as popular as TBs! Mostly, the awards have to *follow* the production of new cultivars in a class and the interest in growing them. Whenever the production of new cultivars has increased to the point where awards become meaningful, awards have been created.

It has to start from the ground up, so to speak. If one thinks there should be a separate class and awards for, say, crested irises, then one needs to get to work with others of like mind and promote them and name new cultivars to the extent that having an award would actually be useful. It may seem like a catch-22, but it's really not. New types of irises can become popular with gardeners and interesting to breeders even before there is an award system for them, but you can't give awards to cultivars that don't exist yet!

Every single awards category today exists because the breeders and gardeners got in there first and produced enough cultivars to make awards meaningful and relevant. This is how our current assortment of categories came about. There is no secret evil clique intent on promoting one part of the genus over others.

Regards, Tom

--- In i*@yahoogroups.com, "Sean A. Zera" <zera@...> wrote:
>
> Tom, you accused those of us claiming a bearded bias of making an
> argument based on an "a priori notion that the bearded irises are
> somehow a small faction of the genus." My point was that clearly they
> *are* a small faction, from a species enthusiast viewpoint. This is
> reality, not a priori, and you did ask the question. Indeed it's
> surely not the best way to break down horticultural awards, nor did I
> claim it was. You've got to admit, however, that for an organization
> that claims to promote all iris, it's pretty biased for hybrids
> derived from half a dozen or so species to take most of the awards
> when the genus offers 300+ to work with. It's not intentional, but
> it's there.
>
> Sean Z.
> Michigan
>
>
>
>
>
> Quoting Tom Waters <irises@...>:
>
> > Hi Sean.
> >
> > I'm afraid I don't see how a numerical count of species tells us
> > much about where awards are needed or appropriate for horticultural
> > varieties.
> >
> > Categories for registration and awards of garden irises do not exist
> > in some abstract botanical world; they exist because they are (in
> > intention, at least) a way for the gardening public to sort through
> > the many varieties in commerce and get a sense of (1) what its basic
> > characteristics are as a garden plant, and (2) how well regarded it
> > is by trained iris judges.
> >
> > The more of a given type there are, the greater is the need for a
> > process of discrimination among them. With hundreds of TBs being
> > introduced each year and destined for places in many thousands of
> > gardens around the world, there is clear value in making the effort
> > to identify the best ones and to distinguish them clearly from, say,
> > BBs, which have quite a different role in one's garden. At the other
> > extreme, there would be little value in say, a three-tiered award
> > system for I. unguicularis cultivars. I think a new one of these is
> > introduced into commerce perhaps every decade or two on average, and
> > they are found in the gardens of a small handful of iris cognoscenti.
> >
> > Somewhere in between these extremes is a cross-over point where
> > there is enough need for a separate class and award system to make
> > it meaningful and viable. It's really up to the enthusiasts for a
> > particular type of iris to popularize the type and generate demand
> > for new varieties. If the demand is there, hybridizers will respond,
> > and eventually it will become clear that a new awards category is
> > needed. This has happened a number of times since the formation of
> > the AIS, as you can see by looking at the history of each award on
> > the Iris Encyclopedia, for example. As my last post explains, the
> > fact that the beardless categories are less competitive indicates
> > that the AIS has had a strong willingness to support these categories.
> >
> > I've been around the iris world for 30+ years, and I've met many
> > judges. I've heard stories of judges who refuse to vote for non-TBs
> > for Dykes Medal, but I've never actually met one. Most of the judges
> > I have known would very much like to see a non-TB win this award,
> > but (except for medians), there is no other category that is grown
> > by enough judges to give it a chance. Most judges grow non-TB
> > irises, but *which* ones they grow tends to be a very regional
> > thing. Here in New Mexico, you can find arilbreds but not many
> > Japanese, Siberians, Louisianas, or PCNs. So if an excellent
> > Siberian is in competition for the Dykes, it simply will not get
> > many votes from this region of the country. This is not because the
> > judges are biased--rather the opposite: they are being ethical in
> > not voting for an iris they have not grown or seen growing over a
> > number of years in another garden.
> >
> > Regards, Tom
> >
> > --- In i*@yahoogroups.com, "Sean A. Zera" <zera@> wrote:
> >>
> >> Well, using a species list I had a hand, lumping as you did the arils
> >> with the beardless, I get almost the opposite effect. Approximately
> >> 15% of iris species are bearded, and 85% are beardless+aril. So yes,
> >> they are a small fraction of the genus. Putting the arils with the
> >> bearded bumps up bearded to 37% of the genus. Comparatively few
> >> bearded species were hybridized to create the current garden beardeds
> >> available, so they're overrepresented even more than they seem,
> >> species-wise.
> >>
> >>
> >> Sean Z.
> >> Michigan
> >>
> >>
> >>
> >>
> >> Quoting Tom Waters <irises@>:
> >>
> >> > It's not at all obvious to me that the AIS or its membership are
> >> > "biased" in favor of the bearded irises. This has been tossed out as
> >> > though it were a simple fact, when actually I think it is very much
> >> > a matter of subjective perspective.
> >> >
> >> > The fact that about half of the classes eligible for AIS are bearded
> >> > classes has been offered as evidence of this "bias". But how so? It
> >> > makes sense only if one has some a priori notion that the bearded
> >> > irises are somehow a small faction of the genus. I don't know how
> >> > one would go about quantifying such a thing.
> >> >
> >> > Here's one way to put the discussion in objective terms. One could
> >> > say that an awards system is "fair" to the different types of irises
> >> > if a new introduction has about the same chance of winning an award
> >> > irrespective of its type.
> >> >
> >> > I did a quick random sample of two hundred registrations and
> >> > introductions from 2009. I found that about 83% were bearded and
> >> > about 17% beardless (I put arilbreds in with the beardless, but
> >> > there were only a couple of them, so it does not much change the
> >> > picture). In other words, there are about 5 times as many bearded
> >> > irises being introduced into commerce each year as all others put
> >> > together.
> >> >
> >> > Now consider that those bearded irises are in competition for six
> >> > awards, and the beardless and arilbreds are in competition for seven
> >> > awards. These means a beardless iris is about 5 times more likely
> >> > than a bearded iris to receive an award in the current system.
> >> >
> >> > How can anyone claim that this state of affairs does not encourage
> >> > the development and recognition of the beardless irises?
> >> >
> >> > In an earlier post, I made this same point a different way, but it
> >> > doesn't seem to have had much effect on the discussion. Maybe this
> >> > attempt at quantification will make it clearer.
> >> >
> >> > If bearded irises represent 5/6 of each year's introductions, it's a
> >> > fair guess that they represent 5/6 of the irises acquired and grown
> >> > by the iris growing public. Is it any wonder, then, that they
> >> > receive the most attention in the AIS Bulletin, and that conventions
> >> > are timed for peak TB bloom?
> >> >
> >> > In my view, the AIS is to be commended for promoting the beardless
> >> > irises *in spite of* their being considerably less popular with the
> >> > iris buying public.
> >> >
> >> > Any discussion of bias that does not acknowledge the large numerical
> >> > disparity in the number of irises being bred and grown in the
> >> > various classes is not very meaningful, in my opinion.
> >> >
> >> > Regards, Tom
> >> >
> >> >
> >>
> >
> >
> >
>



Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index