Australian Problemists >
Ian Shanahan

A. Education and Work

Ian Shanahan (born at Camperdown, NSW, Australia) is a composer, performer (recorders, trombone), improviser, researcher and educator. Ian enrolled in 1980 for the BMus degree at the University of Sydney, completing a double major in music and pure mathematics; he graduated in 1986 with first-class honours and the University Medal. Under the supervision of Anne Boyd, Ian completed during March 2002 a PhD degree (in composition) at the same institution, where he was also a part-time lecturer in orchestration, composition and twentieth-century harmony until 2004. In February 1996, Ian Shanahan was appointed lecturer in performance and composition in the Music Department of the School of Contemporary Arts at the University of Western Sydney.

Ian’s compositions have received several awards and numerous performances both locally and internationally, to widespread acclaim. In addition, Ian has been active as a self-taught recorder player and promoter of new music for this instrument. He has commissioned composers to write for him, and has given many performances, workshops and recordings of their works – as well as lecturing, broadcasting and writing widely about the rich possibilities of the recorder. Ian also enjoys regular forays into certain vernacular, classical, and early musics, having been a member of various classical bands and the Sydney University Orchestra. He has recorded and produced several radio programmes that explore diverse aspects of new music, and written a number of scholarly articles for various publications and guest co-edited Sounds Australian Journal twice. Since 2004, Ian is no longer teaching in tertiary academe. But he still works as a freelance performer, and remains engaged in compositional research while sporadically writing a book about numerical patterns within the Urtext of the New Testament (gematria) at the same time as gradually archiving as much of Australia’s chess-problem heritage as possible.

B. Chess Problems

Ian was taught chess by his elder brother Chris in 1972 or 1973, and soon began to solve on a weekly basis the “White to play and mate in 2” chess problems then republished within Sydney’s Sunday newspaper, The Sun-Herald. The chess column therein was the crucible for firing enthusiasm in chess problems not only for Ian, but for Peter Wong and Geoff Foster as well; all three names regularly appeared among the weekly lists of prize-winning solvers. However, Ian did not start composing problems himself until 1977 (having acquired Brian Harley’s book Mate in Two Moves and Kenneth S. Howard’s various textbooks devoted to chess problems); apart from this – and the situation was exactly the same for Wong and Foster – he had no expert ‘problem mentor’ to guide his early compositional attempts. His first original problem was published in 1979, the year after he joined the British Chess Problem Society on the recommendation of Bob Meadley (then editor of the Problem Corner in Chess in Australia), whose magazine The Problemist thus became, and still remains, his primary guide in the problem art and his main compositional outlet. Ian also edited the “Problem Billabong” column in Australian Chess (now known as Australasian Chess) from 2003 to 2007.

Ian’s favourite problem genres are the orthodox two-mover and series-movers (often combined with other unorthodox [‘Fairy’] elements), though nowadays he focusses mainly upon the latter. The following half-dozen problems therefore reflect these predilections:

1. Ian Shanahan
The Problemist, Jul. 1998
~ To Michael Lipton ~
Mate in 2

Set: 1…d5 2.Qxe5≠.
1.Qxd6! (>2.Qd4 [A], Qd5 [B], Qxe5 [C])
1…Sb7 2.Qd4 [A], Qd5 [B], Qxe5 [C]≠.
1…f6 2.Qd4 [A], Qd5 [B]≠.
1…Qxg4 2.Qd4 [A], Qxe5 [C]≠.
1…Sb3 2.Qd5 [B], Qxe5 [C]≠.
1…Saxc4 2.Qd4 [A]≠.
1…Sac6 2.Qd5 [B]≠.
1…gxf5 2.Qxe5 [C]≠.
1…Sf3 2.Bd3 [D], Bf3 [E], Rxf4 [F], Qxf4 [G], gxf3 [H]≠.
1…Sec6 2.Bd3 [D], Bf3 [E], Rxf4 [F], Qxf4 [G]≠.
1…Sexc4 2.Bd3 [D], Bf3 [E], Rxf4 [F]≠.
1…Sd3 2.Bd3 [D], Bf3 [E]≠.
1…Sxg4 2.Bd3 [D]≠.

This ‘traditional’ single-phase problem – to the best of my knowledge uniquely among all two-movers! – parades a blend of three of my favourite two-move themes: Dalton theme II (White unpins a black man, which then fires a battery pinning its unpinner); total primary combinative separation (the three primary threats appear as mates in every possible combination, each one after a single black defence); and progressive separation or mate reduction (six new mates are whittled down successively to just one). There are also eight ‘levels of intelligence’ of black moves – a kind of ‘octary correction’. Some readers may find the algebraic pattern-play of this composition to be rather radical – until it is remembered that such ‘thematized multiples’, as embodied by the latter two themes, date back to the late 1940s. I was somewhat shocked (and disappointed) that this singular problem received no tourney honours whatsoever; its dedicatee, the famous British International Master Michael Lipton, thought it deserved 1st prize!

Now to an altogether more modern style of two-mover:

2. Ian Shanahan
The Problemist, Mar. 2001
~ To David Shire ~
Mate in 2

1.Sd4? (>2.Qxb5) 1…Rh5! (2.Qd1?)
1.Sa1? (>2.Qd1 [X]) 1…h1(Q)! [a], Rh3! [b]
1.Se3? [A] (>2.Qd1 [X])
1…c3 2.Qb3≠.
1…h1(Q)! [a]
1.Se1? [B] (>2.Qd1 [X]) 1…Rh3! [b]
1.Qd1? [X] (>2.S~)
1…h1(Q) [a] 2.Se1 [B]≠.
1…Rh3 [b] 2.Se3 [A]≠.
1.Bb6? [E] (>2.Qa8 [Y]) 1…Rh8! [c], Rh7! [d]
1.Bc7? [C] (>2.Qa8 [Y]) 1…Rh8! [c]
1.Bd8? [D] (>2.Qa8 [Y]) 1…Rh7! [d]
1.Qa8! [Y] (>2.Bc7 [C], Bd8 [D], Bb6 [E])
1…Rh8 [c] 2.Bd8 [D]≠.
1…Rh7 [d] 2.Bc7 [C]≠.
1…d5 2.Bb6 [E]≠.

Decisions, decisions… Which battery to set up, vertical or diagonal? And which piece will move first, the front or rear? After playing through the logical evolution of tries, we discern:

(i) the (nowadays quotidian) Banny theme, exhibiting the ‘algebra’ {Tries: 1.A?/B? 1…a!/b! 1.Key! (or 1.Try?) 1…a/b 2.B/A ≠} through the commonplace mechanism of battery-formation and -play, yet here in a rarely-seen doubling of the pattern; in conjunction with

(ii) try/key + threat sequence-reversal, doubled – i.e. Qd1 [X] and Qa8 [Y] serve as both first-moves and post-try threats within the formal plan {Tries: 1.A?/B? (>2.X) Try: 1.X! (>2.A, B, …) Tries: 1.C?/D?/E? (>2.Y) Key: 1.Y! (>2.C, D, E)}, while the tries 1.A–E become threats; the play after 1.X? and 1.Y! also displays total change (i.e. defences and mates are all altered); and

(iii) the partial Fleck theme, wherein the 3 post-key threats (>2.C, D, E) are separated into individual mates by specific black defences – less careful ones of which, alas, lead to multiple mates (hence the Fleck effect is just “partial” instead of total). Notice too the funktionwechsel (exchange of rôle) between the white knight and bishop in guarding b4 throughout the various phases. The black rook is a valiant defender indeed!

3. Ian Shanahan
Australian Chess, Jan. 2003
Mate in 2

1.Bc5? (>2.Qxd6, Qd4)
1…Rh6 2.Qd4 [A]≠.
1…Sxf5 2.Qxf5 [X], fxe4 [Y]≠.
1…dxc5 2.Rd8≠.
1.Sc4? (>2.Qxd6)
1…Rh6 2.Bc3 [B]≠.
1…b3 2.Bg1 [D]≠.
1…Sxf5 2.Qxf5 [X]≠.
1…Sc5 2.Sb6≠.
1.Be5! (>2.Qxd6)
1…Rh6 2.Sb3 [C]≠.
1…b3 2.Sf1 [E]≠.
1…Sxf5 2.fxe4 [Y]≠.
1…dxe5 2.Qd8≠.
1…Be6 2.Qxe6≠.

The array on the d-file – with tries by alternate pieces thereon – constitutes a half-battery (a configuration seen ubiquitously throughout the 1960s and ’70s), here showing five changed mates, A–E, after 1…Rh6 and 1…b3, between the 1.Bc5?, 1.Sc4?, and 1.Be5! phases. Across all three phases after 1…Sxf5, one can also discern the less run-of-the-mill Mäkihovi theme: in a set- or try-phase, a black defence allows two white mates X and Y (i.e. a dual) which are forced singly in further try- or post-key play. (The white pawn f3 and black knight g3 that carry this theme were added to the position towards the very end of the composing process, almost as an afterthought – but one that is definitely worth the [minimal] extra material!) There are twelve mates in all (including threats), in a style that articulates traditional two-move strategic elements – here primarily line-effects – within a fairly modern multiphase carapace!

4. Ian Shanahan
U.S. Problem Bulletin, Nov. 1994
2nd Prize
~ To Geoff Foster ~
Series-helpstalemate in 18
(b) Qh3 to h2

(a) 1.Rh2 2.Bh1 3.fRg2 4.Bf2 5.Rg1 6.Bg2 7.gRh1 8.Kg1 9.Bf1 10.Rg2 11.Kh2 12.Bg1 13.Rf2 14.Qg2 15.Kh3 16.Rh2 17.Qh1 18.g2, Sf5=.
(b) 1.Bh3 2.Qg2 3.Rh2 4.Qh1 5.Bg2 6.Rh3 7.Bh2 8.Kg1 9.Bf1 10.Qg2 11.Kh1 12.Bg1 13.Bg1 14.Qh1 15.g2 16.Kg3 17.Rh2 18.Kh3, Sf5=.

The motto of this problem – which took well over 200 hours to compose ‘by hand’, without computer assistance – is “Parliament House”: not only is its shape reminiscent of the profile of Australia’s Federal Parliament building, but the units therein likewise shuffle about until complete immobility ensues! Geoff Foster (the master of such follow-my-leader [FML] chess problems, akin to sliding-block puzzles) was the impetus behind the creation of this complex twin, which shows an 18-fold FML-chain doubled, and justifiably is its dedicatee: towards the beginning of our friendship, Geoff sent me the sketch-material and compositional methodology for his own FML precursors, which proved inspirational to me; his constructional techniques, derived from Game Theory, later formed the basis for a set of articles on the subject within The Problemist Supplement. The judge of the tourney within which “Parliament House” competed, the German Grandmaster Hans Peter Rehm, had this to say about it:

“Twins (or multiple solutions) in seriesmovers are rare, and extremely rare are those in more than 10 moves. The twin not only doubles but (at least) squares the value of the invention (if the thematically related solutions, as here, are varied enough). The solver admires how it has been possible to obtain two very different precise sequences of exactly the same length. If one studies the mechanism one finds several interesting permutations of pieces: in the final [topologically identical] stalemate positions rooks have interchanged their place; both twins start with a Platzwechsel (in a. [Rf2/Bg1 (after two moves)], in b. Qh2/Rg1). In a. compare the positions in the diagram and after the 15th move: Platzwechsel of Rf2 and Rh1 and cyclic Platzwechsel of K, Q and B. In b. compare the positions after the 2nd and 12th move: cyclic Platzwechsel of K, R, Q, and B.”

Since there are several routes forward (and that can be retracted from the final position), I envisage that “Parliament House” would be tough to solve; indeed, the diagram position admits two non-trivial tries in 19 moves. One must not play g2 too soon! And besides merely perfecting the ‘architectural’ shape, the black pawn on h6 prevents a cook ending with 13.Qxh4+, Kxh4 =. (Likewise, the white pawn on g4 stops short-circuits whereby men exit the ‘cage’ only to re-enter it later.)

5. Ian Shanahan
The Problemist, May 1995
4th Hon. Mention = (1996)
~ To Arthur Willmott ~
Series-selfmate in 20

1.b8(B) 2.a8(Q) 3.Qa7 4.Qxc5 5.a7 6.a8(R) 7.Ra6 8.Rf6 9.Ke6 10.Be5 11.Qd6 12.c5 13.c6 14.c7 15.c8(S) 16.Se7 17.Sg6 18.Sf4 19.Sh5 20.Sg7+, Sxg7≠.

This problem was composed in five minutes flat (after I had been thoroughly analyzing some series-movers by the late Brian Tomson, the mechanisms of which were still swimming around in my head)! It shows the omnipresent AUW (“Allumwandlung”: all four promotions) ending with an ideal mate in an economical setting that requires only one capture, as well as white-king and -queen hesitation, critical play (moves 8–11), and a long, accurate white-knight trek thrown in for good measure (this last feature imparts some degree of originality to an otherwise hackneyed concept). Apparently, this problem is also quite difficult to solve (Bob Meadley was tearing his hair out over it for many hours until the penny finally dropped, and it defeated some of The Problemist’s best solvers entirely!); there is a ‘cook-try’ sequence in 21 moves (a false trail, just one move too long, with variable move-order) – ending with the white king on c8, a white queen on g7, a white pawn on c7, a white minor piece on b8, and 21.Sd6+, Sxd6≠ – that might have seduced the unwary. (Observe in this alternative scenario that the white pawn on d5 cannot move in order to facilitate more rapid access for the white king to reach c5 – since then 21.Sd6+ would be nullified.)

6. Ian Shanahan
The Problemist, Nov. 1997
Series-selfmate in 10

1.d8(B)! (d8(Q)?) 2.Bg5 3.Bd2 4.0-0-0 5.Rg1 6.Rg6 7.Kb1 8.Ka2 9.Rd6 10.b4+, cxb4 e.p.≠.

Inspired by one of Brian Tomson’s lovely series-selfmates, this was my very first problem – the instigator for an ever-growing corpus amongst my series-movers – to demonstrate the now-fashionable Valladao task (i.e. promotion, castling, and e.p.-capture), without white captures or ‘cookstoppers’, plus two shields of the white king that help to control the move-order wherein only the thematic units are mobile. The black knight on a5 (which must not be captured) determines precisely the route of the promoted white bishop. It’s a pity, however, that the black queen is not used to self-block the black king – a breach of economy – but she does have other multiple functions, such as ensuring that both White’s promotion and the promotee’s route to d2 are unique. Appreciate, too, how castling both accelerates the white king’s access to his destination while acting as a clearance manoeuvre for the white rook to gain g6 in just two moves. There is also an 11-move ‘cook-try’ (whose move-order is not fixed): 1.Se5 2.d8(Q) … 4.Rd3 … 8.Ka2 … 10.Ra4 11.b4+.

On a personal note, this composition indirectly sparked my ongoing friendship with the British problemist Mark Ridley, who is an avid collector of Valladao-task problems: we are in the process of writing an extensive joint article devoted to the Valladao task in series-movers, which will also include a theme tourney for new examples.