Jean Rene Vernes' "Canape Majeure d'Abord"

Having mentioned in passing Jean Rene Vernes' (JRV in the following)
"latest" book (well, it's from 6 years ago, but I don't think he's
written any since;-), "Bridge Distributionnel" (BD), and the new
simplified "Canape Majeure d'Abord" (CMA) system it describes, I've been
asked in email to summarize it on r.g.b. If you're not interested in
obscure systems (even though propounded by a major theoretician of
modern contract bridge), by all means skip this post!-)

The following text is covered by Creative Commons Deed, Attribution 2.5
( -- IOW: reproduce and make
derivative works if you want, but attribute the work to "(c) Alex
Martelli 2006, summarizing ideas by Jean Rene Vernes". Ideas, of
course, cannot be subjected to copyright, so you're legally free to
adopt and/or adapt any of the following ideas (as opposed to, my text
expressing those ideas!-) without any obligation to acknowledge their
source in any form whatsoever, though many people consider it courteous
to do so anyway. No warranty expressed or implied!-)

In the following I use many common bridge abbreviations: m=minor,
BAL=balanced, UNB=unbalanced; S/H/D/C for the suits (H is ambiguous as
it may also mean "High-card points" but context hopefully clarifies).

1. hand evaluation

JRV's ideas were extensively discussed and explained in an earlier work
(L'Evaluation des Mains au Bridge) cowritten with Bernard Charles, with
abundant statistical and bridge-detailed analysis. Summarizing:

H: 4-3-2-1 basis. ONLY the following adjustments:
-1 for a bare honor (fully including the Ace)
-1 for 2 bare honors _without_ the Ace (KQ, KJ, QJ)

DH: H, plus 3-2-1 for shortness, +1 for a decent 6-card suit, +2 for a
decent 7-card suit. However: 4432 shape is actually worth 0, 5422 is
only worth 1; 4441 is not worth a full 2; voids are worth more than 3:
553 shape in particular is worth closer to 4 than to just 3.

S: similar to DH, but no values for side lengths; 4432 is worth 0 if
raising on 3, but 1 if raising on 4; voids are worth 4 with 4+ trumps.
If responder knows that the pair has 9 trumps he must add +2 S; if
responder knows the pair has 10 trumps, +3; if 11, +4.

JRV has many other interesting and useful ideas on hand evaluation, most
strikingly the "mixed points" (M) for minor 1-suiters, but they're not
used as primary guidelines in the CMA system.

2. the "basic" system on which CMA is founded

JRV presents CMA in terms of differences from the "Standard French" (SF)
system: 5-card majors, strong NT, weak 2 in the majors, "Benjamin-like"
2m (2D gameforcing, 2C artificial and strong but below GF). In
particular, in BD he suggests playing the following opening bids just
about like one does in SF: 1S and 2S, 1N and 2N, 2C, 2D.

Small variations: 1NT is 16-17H (not 15-17), so a Stayman response
followed by any rebid by responder is GF; 1NT may be a 6322 with minor
length and good side stoppers (rebids 3m on Stayman) as well as a 5332
with 5 hearts (but normally not with 5 spades, unless very bad ones).

2C includes hands with 19-21DH, exactly 5 hearts, and less than 4
spades, in addition to the "normal" hands of 22-24 DH (UNB) and 22-23H
(BAL) -- and is less restrictive than SF about shape and suit quality
for the 22-24 DH UNB hands. Vice versa, 2C and 2D exclude game-forcing
UNB hands with exactly 4 cards in spades (4441 or 4S+longer suit), see
below the 1C opening bid.

JRV has his favorite criteria and continuations over these opening bids,
too, but he believes the advantages of his approach versus SF's are
minor and he's willing to give that up in the hope of a more widespread
adoption of CMA.

JRV *IS* insistent on opening 2S with a good 5-carder and a side
singleton; he's proven rather strongly that the "traditional"
distributional requirements (6-carder, for 2S) are too heavy... or
rather misplaced: 6322 is fine for 2S and 5322 would be bad; JRV's point
is that 5431's (and 55 hands even more so) are more akin to 6-carders
than to 5322's -- side suits also matter in determining distributional
safety, not just the main suit. ((Similarly, it's wrong to believe that
a 4333 with a 4-card fit gives distributional safety at the level of 3
when raising partner's 5-cards suit -- 4432 with the same fit would be
fine, but the bid is beyond distributional safety w/4333)).

However, these issues wrt the exact definition of preemptive opening
bids and "distributional security" are only marginally "systemic".

3. CMA-specific opening bids

3m: 6-7 cards, one-suiter, 14-16 DH (usually 10-13H)
2H: any 6H, or 5 good H and a side singleton, never 4+S, 13-15 DH
(usually 10-13H)
1H: 4+H, less than 4S, 12-19 H / 13-21 DH; unsuitable for opening
1N (16-17H BAL), 2H (see above) or 2C (5H, 19+DH)
1D: no majors, 12-19 H / 13-21 DH; unsuitable for opening 1N or 3m;
can also be 22-24DH with a minor 2-suiter
1C: exactly 4S, 12-19 H / 13-21 DH unsuitable for opening 1N; can also
be (always w/ exactly 4S) an unbalanced hand with 22+DH.

JRV also often uses single-letter notations for hand strength: A=13-15DH
(or 12-13H BAL), B=16-18DH (or 14-15 H BAL), C=19-21DH (for BAL, C1 is
16-17H, opened 1N, and C2 is 18-19H) -- and similarly on up for D and E
(D1 and D2 for BAL). The average/typical UNB hand is worth 2 D points
(so A=13-15DH normally corresponds to 11-13H) but a few are worth just 1
D point (5422, 4441) and many are worth 3 or more (6331, 5521, ...), so
the H strength of the UNB hands' strength-zones can vary a bit.

Responder often relays over 1C-1D-1H by bidding the first step (10+H
over 1D and 1H, but promises no strength over 1C) and opener's rebid
shows his strength zone: force C often rebids 3C-D-H, force B often
rebids 2C-D-H, force A always rebids 1N (after opening 1D or 1H; always
1S after opening 1C) which just shows minimum strength w/o further
distributional information for now.

4. 2H opening bid and continuation

Chance of color is nonforcing; a 2N relay is the only forcing bid,
opener rebids naturally (a side minor, or 3H=1-suiter, thus 6+H)

5. 3m opening bids and continuation

3D on 3C is a relay (opener bids 3N=no singletons, 3S=H singleton,
3H=either D or S singleton); other responses are natural, forcing to 4m
if under game.

6. 1H opening bid and continuation

1S is a relay (any hand of 10+H); opener rebids:
1N = any shape, force A
2C-2D: canape (4H/5m), force B
2H: 5+ hearts, force B
2S: BAL, 4/5H, force B (can be 1=4=4=4, 14-16H)
2N: BAL, 4/5H, force C2 (18-19H); can be 1=4=4=4, 17-19H
3C-3D: canape (4H/5m), force C
the rebids above 1N are GF. On 1N, responder may relay again (2C) with
12+, and opener rebids:
2D or 3C: canape
2H: 5H
2S: BAL (can be 1=4=4=4, 12-13H) [2N too: distinction unclear]

Further relays may provide other distributional information, etc etc.
However, BD's focus is on simplicity, so bidding is mostly natural in
these continuation sequences (no artificial steps &c, beyond responder's
1st step being just a "tell me more" relay) with a very few exceptions:
e.g. on responder's relay over "BAL or 1444" 2S bids, opener's rebid of
3S shows the 1444 shape).

The 1N response denies H fit and shows 6-9H; opener often passes
(including, e.g., with 5422 and up to 14H or even 15) but should rebid
if he has 16+ (rebids don't necessarily promise strength, could be
2-suiters which cannot stand NT). 3m shows 19-21 DH canape, just like
on a relay, but of course is not forcing, just invitational.

Raises are normal but limited to max 9H (relay first with 10+H): 4H=6H,
or 5H and some singleton; 3H=5H, or 4H and some singleton, 8-9H; 2H, 4+H
and 6-9 H (just 6-7H if has a side singleton or 5H).

Other responses: 2S=6 cards, 6-9H; 2m=same (but could be 5 cards and a
side singleton).

7. 1D opening bid and continuation

1H is a relay (10+H); opener rebids:
1S = minor 2-suiter, force B or C
responder relays 1N, opener bids the longer suit, 2m=B or 3m=C
1N = force A (must be BAL or minor 2-suiter: 1-suiter would open 3m)
if responder (w/12+) relays 2C, opener bids 2H=BAL, or with a
minor 2-suiter the longer suit, 2D or 3C
2C-D = one-suiter (6+ cards), force B
2H = BAL, force B (14-15 H)
2S = minor 2-suiter, force D (22-24 DH)
2N = BAL, force C2 (18-19 H)
3C-D = one-suiter (6+ cards), force C
(similar logic as after a 1H opener; if responder relays on a BAL rebid,
opener bids a 5-card minor, a 3-card major with a 4432 [doubleton other
major], NT with 4333 [either 4-card minor]).

1S shows 5+S, 4-11 H, and is not forcing; opener (besides passing or
making natural rebids) may rebid 2H, artificial, to show a minor
2-suiter, force B-C.

1N shows 6-9, similar logic as after a 1H opener. 2C-2D like on 2H
(6cards, or 5cards and a major singleton, 6-9H); 2H-2S = 6 cards, 6-9H.

8. 1C opening bid and continuation

1D is a relay (but promises no strength); opener rebids:
1H = 4+H, force C _or more_ -- BAL, 4441 (either m singleton), or
canape 4S+5H -- responder normally relays 1S w/5+H, and opener:
1N = BAL 16-17H; 2m=4 cards in 4441 16-17H
2H/3H/4H = 4S, 5H, respectively force C/D/E
2N = BAL 18-19H; 3m=4 cards in 4441 18-19H
3N = BAL 20-21H; 4m=4 cards in 4441 20-21H
5m=4 cards in 4441 22+H
[note: JRV prefers to avoid opening NT with 4-4 in the majors]
1S = force A
if responder (w/12+) relays 2C, opener bids 3D/H/S w/4441 and
12-13 H (suit just above the singleton), 2S or 2N BAL 12-13 H
(criteria to choose between 2S and 2N unclear in BD), 2D, 2H
or 3C as canapes with 13-15DH
1N = BAL or 4441, 14-15 H (on a 2C relay, 3D/H/S show 4441,
other rebids are natural and confirm a BAL hand, 2N=4333);
may be 16H if 4441 w/H singleton specifically
2C/D/H: canape, 4S+5 other suit, force B
2S = forcing, 4S+5 any other suite, force D or more
2N = BAL 18-19H (denies 4H), may be 17H w/4=1=4=4 specifically
3C/D/H: canape, 4S+5 other suit, force C

Responses much like over other opening bids: 2/3/4 S "raises" like the H
ones over 1H, 2C/D/H with 6-9 H and a 6-card suit. A response of 1S
shows 0-7H with 3 spades, or 0-7DH with 4 spades, and strongly urges
opener to pass (if opener rebids he shows at least 17-18H). A response
of 1H shows 4 cards in hearts and 6-11H, denies 4+ spades, but is a
1-round force; opener rebids like on a 1D relay, except that, with a H
fit, he raises to 2H, 3H or 4H depending on strength.

A 1N response to 1C artificially shows exactly 5 hearts and 6-9H.
Responder should prefer 1N or 2H to 1H if he qualifies for these more
specialized responses; on 1N, opener can pass, preference (or jumpraise)
H, make a minor canape (strong, NF but invitational, if a jump), raise
to 2N or 3N with BAL (or other hand suitable for NT play) and 17-18H or
19+ H respectively; or, lastly, bid 2S, force D or more, like on a 1D
relay (here forcing _to game_ since responder has 6+H, of course). Note
that this 2S rebid is also available on all other responses, with the
same meaning (except that on 1S it requires E+, as opener can jump
canape to 3 of a suit with force about D, since a canape to the level of
2 already shows force C or thereabouts).

9. slam bidding

It's based on a 4C bid (by responder only, after relays) asking for
controls "in bulk" -- A=2, K=1, don't count a bare K (do count a bare A
fully). Opener with force A bids 4D=2 controls, 4H=3 controls etc; with
force B, 4D=3 controls, etc; with force C, 4D=4 controls, etc. When
opener's last response to a relay is 4C, 4D by responder plays the same

Responder may continue with 4N asking opener to bid the cheapest suit
where opener has *** no K, no Q, and no singleton nor void *** (opener
replies are not completely detailed in BD, particularly for the case
where opener has already shown a specific singleton or void in the relay
phase). After that, responder must place the contract.

10. when opps overcall

Responder's double shows 10+H and takes the place of the relay (no need
for negative double since the opening bid never "hides" opener's majors
of 4+ cards); other calls by responder deny 10+H (with 10H or a bad 11
responder may decide to pull in a notch and eschew the relay, depending
on how he likes the prospects for NT play in particular, if that strain
appears to be a likely landing place from his POV). Bids of 1 of a suit
are like without the overcall; 1N=8-10H, natural; 2 new suit=8-10H,
natural, NF of course. Jumps in new suits are semi-weak,
semi-invitational, depending on what alternatives were open to responder
w/that suit, but always less than 10H. A cuebid in overcaller's suit is
GF, usually with a singleton or void, but possibly just xx and a strong
side minor, looking for NT.

Agreements over opps' TO double are similar, with the redouble being the
artificial relay 10+. The bid that would usually be a relay becomes
natural, 5 cards, NF, much like 1S over 1D, see above.

On overcalls at the level of 3, double still shows 10+H but also denies
a singleton, to make it easier for opener to convert to penalties; bids
in a suit are invitational but NF.

On interventions of 4th hand after our opening and relay: NT bids show
the same strength (e.g: 1NT=force A after 1D-(P)-1H-(1S)), but also a
strong stopper in their suit (ideally a double stopper for 2NT) --
opener passes with force A without a stopper in their suit (or if their
bid's 1NT or above) -- with B or more, enough for game, opener cuebids
their suit to show that he lacks a sufficient stopper (and that he would
otherwise aim towards NT).

Opener's double shows 4+ good cards in the suit they bid, if that's
possible, and proposes a penalty; when opener cannot have the suit they
bid (1D opener doubling a major, 1H opener doubling spades), it's a
takeout double instead.

11. my evaluation of CMA, and some suggestions

CMA as presented in BD seems a tad oversimplified to me, particularly
the continuations when responder keeps relaying, and slam explorations
-- how the system would perform in such situations appears to depend
strongly on the luck of which suits happen to be held and/or
interesting, while at the same time the meaning of some of opener's bids
remains unspecified. "Relay breaks" by responder are also seriously
underspecified. I find it somewhat funny that responses of 2N and 3 of
a new suit are "nonexistent", too;-).

On the plus side, these simplifications should make the system rather
playable "at the table". Anybody interested in trying CMA out beyond a
couple of experimental sessions would probably be well advised to reach
better definition for all sequences, and adapt the system's principles
to the opening bids that CMA currently suggests you play "just like in
SF". In particular, treating the 2S opening bid (a normal weak 2, but
may be good 5 spades with a side singleton) much like 2H (just weaker --
say 10-12DH for 2S, while 2H is 13-15DH) seems easy (and this agreement
was indeed specified in the previous versions of JRV's "Majeure d'Abord"

All the following suggestions are mine alone and JRV bears no
responsibility or guilt for them...:-).

The 1S opening bid could usefully be limited to 13-18 DH (if hands with
5H can bid 2C-then-2H with 19-21 DH, it seems to me that hands with 5S
should be able to do the same!), except when the 19-21 DH are due to a
wild shape with few H (in which case a high rebid over a 2C relay will
show the exceptional hand, see below).

Over 1S, we'd have the usual CMA arrangement: a weak NT response (6-9H
like on 1D and 1H), a 2C relay (all hands of 10+H), and normal raises
(with 9-H). I'd like to treat both 2D and 2H responses as showing H,
much like 1H/1N/2H responses over 1C -- it makes sense to devote more
bids to major suits when they need them (and indeed that's the very
essence of the system, even its name, after all:-). E.g., 2H 6-9H with
6 hearts, 2D 6-11H with 5+H (opener if minimum bids 2H with 2 or more
hearts, 2S with H shortness, 3H invitational wih 3 or more hearts;
2N-3m-3S natural and strong; responder rebids naturally and
invitationally over 2H/2S with a 10-11 hand) -- this would allow, after:
1S 2C (10+)
2D (force A -- I think it should deny 4+ hearts or 6+ spades)
responder's relay of 2H as GF. I also believe that:
1S 2C
should show force A and 6 spades (NF, but any responder's rebid,
particularly a 2N relay asking for a 4-card suit, is GF); and:
1S 2C
should show 4+ hearts but ambiguous force (AB) excluding only a 55 shape
of force B (which would rebid 3H on 2C) and a 64 shape of force A (which
I think should rebid 2S on 2C) -- still a 1-round force.

On 2H, responder with just 10-11H could bid 2S (could be doubleton) or a
natural, NF 3C/D (6 cards) or 3H/S (natural, INV raises); 2N would be
the GF relay, on which opener would finally show his strength (3C=13-15
DH 5431, or 5422 12-13H; 3D=16-18 DH 5431, or 5422 14-15H; 3H=55 force
A, 3S=64 force B, 3N=5422 16-17H, 4m=4-card suit in 5440 force B, 4H=65
force B, 4S=74 force B).

Other rebids by opener over 1S-2C, all GF:
2N BAL (5332) 14-15 H
3C/D 4+ cards force C (16-18 DH)
3H 5-5, 16-18 DH
3S 6-card 1-suiter 16-18 DH
3N BAL (5332) 16-17 H
4C/D/H: 6-5 or more, 19-21 DH (i.e., about 13-16 H)
4S: 7 or more cards, 19-21 DH (about 13-16 H)

E.g., AKJxxx-KQJxx-x-x should of course open 1S (not 2C) despite the
19DH, and would show the exceptional shape and strength with a 4H rebid
(may be worth it even over a 1NT response, actually, just hoping partner
has an A or the SQ to make it playable;-).

An immediate 2NT response over 1S (and I believe over other openers too)
should show both minors, 5-4 or 5-5, with about 10-11H (NF, but it's UNB
and has NO implications of major stoppers!); 3C/D/H over 1S (and again,
I believe, similar 3-level responses on other opening bids) should show
7 cards (possibly 6 good cards, particularly with a side singleton) and
the usual 6-9H, basically like the corresponding opening bids (since a
1NT response is natural and often passed, you don't really want to make
it with this kind of hands!-).

The rough slam-investigation approach might be usefully complemented by
some kind of "spiral scan" arrangement.