Canary Club

This is an old revision of the document!

---

<p> The Canary system was devised by Paul Heitner and John Lowenthal sometime in the late 50's or early 60's. There is a high level description of the system in <i>The Encyclopedia of Bridge</i>.

<p> The system is really Roman

with a strong 1C opening
rather than the Roman weak 1C
opening. One of the drawbacks to Roman C
was the necessity to create the suggestion of strength by bidding two suits
rather than bidding and rebidding one suit. In addition there was no way to
show a C one suiter
with the opening bid - one always had to bid a shorter suit first. As a consequence,
there frequently were opening bids in 3 card suits.

<p>&nbsp; <p> Canary avoids that problem altogether. In fact there is no bid in the system

of a suit less than 4 cards in length unless the bid is pre-defined as artificial.
There is no illusion of naturalness as there is in Roman C
or in so-called natural systems.

<p>&nbsp; <p> A second feature of Roman C

was the use of a step response to 1D/1H/1S as an artificial
negative. This is part of the description of Canary C
which appears in <i>The Encyclopedia of Bridge</i> too. It has also been a basis
of Canary C as described
in earlier versions of this document.

<p>&nbsp; <p> As of version 3, we will substitute 1N as the first relay response. This serves

two purposes:

<ol>

<li>This puts us more in line with what the ACBL considers legal. The 1N response
  to 1D being treated
  as forcing is unique to this system, I believe, but probably causes no &quot;legal&quot;
  problems.
<li>Theoretically it probably is superior as well since it causes the hand which
  is more likely to be balanced to be declarer.

</ol> <p>&nbsp; <p>There is a problem introduced with this change. Auctions such as 1D

1. 1H - 1N (which showed

a D suit with a H

suit at least as long as the D
suit) and 1H - 1S
- 1N (which showed a H
suit with a S suit at
least as long as the H
suit) served as a way for opener to show that he didn't have "true"
reversing values in the old method. There no longer is a means to show the distinction
between reversing values and weaker hands. Experience will show whether this
is a problem or not. <a name=basis>

<p>&nbsp;