The Batucada System: Brazilian

Companion to Chapter 10: Brazilian. This page states the counterpoint rules of samba (and its bossa distillation) explicitly. Everything below is summarised from the published scholarship listed under Sources — prior research restated in Poly’s terms, not original theory. See About This Guide for what this guide is, what it is not, and what it leaves out.

The System

Samba’s ensemble is organised by three interlocking referents rather than one. The surdo pair supplies the metric floor as a dialogue: the first surdo answers on beat 2, the second marks beat 1 — and crucially, samba is beat-two weighted: the deep, sustained stroke falls on 2, inverting the Western downbeat instinct. Above it, the tamborim’s telecoteco pattern functions as the timeline — Sandroni analyses it as samba’s characteristic 16-pulse paradigm (his “Estácio paradigm”), a contratempo figure that took over from an older, more onbeat paradigm in the late 1920s (Sandroni 2001). Underneath everything, the caixa supplies an unbroken sixteenth-note carpet with an internal accent pattern. Every other voice — agogô, repinique, chocalho, cuíca — is defined against these three strata (Fryer 2000 traces the system’s African ancestry).

The feel layer is unusually well measured: Brazilian sixteenths follow a systematic long-short-short-long timing profile with position-dependent amplitude, stable across players and tempi (Gerischer 2006; Naveda et al. 2011) — a signature that neither straight grids nor swing percentages reproduce.

The Rules

1. The surdo dialogue is fixed, and beat 2 is the heavy one. Second surdo strikes 1 (often damped), first surdo answers 2 (open, ringing, louder). A third surdo, where present, adds syncopated pickup strokes into the pair. This alternation never varies during a section; it is the floor everyone dances on.

2. The telecoteco is the timeline, and it is contratempo. The tamborim/agogô stratum carries a fixed 16-pulse pattern that leans off the beats — its accents surround rather than strike the pulse (Sandroni 2001). Like clave (its structural cousin), it has directionality: the pattern’s syncopated and grounded halves must sit consistently against the two-bar sense, and parts derived from it inherit that orientation. It never mutates.

3. The caixa never stops. Continuous sixteenths, ghosted, with a repeating internal accent figure — the connective tissue that makes the ensemble roll. Gaps in the caixa stratum are section events (breaks commanded by the repinique), not texture.

4. One voice commands. The repinique (or the mestre’s whistle) has the call role: entrances, breaks (paradinhas), and section changes are cued by its signal figures. Improvisation elsewhere is minimal — a batucada is a machine of ostinatos with one speaking voice, closer to funk’s riff-lock than to jazz.

5. Syncopation is around the beat, not against the meter. Brazilian syncope displaces emphasis to the sixteenth before or after the beat (the characteristic tied-note figures), while the surdo floor keeps the beat itself unambiguous. Free polymeter over the samba grid is not idiomatic; all voices share the 16-pulse frame (same-grid layering, as in the Balkan and Manding models).

6. The feel is long-short-short-long. Within each beat’s four sixteenths: first slightly long, middle two compressed, fourth slightly long again, with the profile’s depth varying by instrument (Gerischer 2006; Naveda et al. 2011). It is a timing signature, mild enough that naive listeners call it “straight but alive.” Swing parameters approximate it poorly — they displace only alternate notes, where the profile is a distribution across all four. Poly expresses it as a per-lane subdivision profile: the ratios themselves, normalised so the bar keeps its length, which is why they survive a tempo change where a fixed millisecond offset would not.

7. Bossa is the system distilled, not simplified. One displaced cell (the E(5,16) bass figure, or the guitar’s thumb-and-fingers reduction of surdo + telecoteco) against an even pulse, at whisper dynamics. The rules survive the reduction: beat-2 weight in the thumb, contratempo lean in the upper voice, continuity in the ride/brush stratum. What disappears is mass, not structure.

Construction

1. Surdo floor. One lane, 8 steps at 1/8, hits on beats 1 and 2 positions with velocity asymmetric — beat 2 stronger (Rule 1). (Two lanes at different pitches if you want the true pair.)
2. Telecoteco stratum. 16 steps at 1/16, timeline mode with the tamborim pattern (seed from E(5,16) or E(7,16) and adjust to the contratempo figure — the exact pattern varies by school; what matters is fixed, offbeat-leaning, never mutated) (Rule 2).
3. Caixa carpet. 16 steps, 13–16 hits, low velocity, ghost 50+, accent mask carrying the internal figure (Rule 3).
4. Colour voices. Agogô (E(5,16) rotated), chocalho (dense, quiet), each on the shared 16-grid (Rule 5).
5. The commander. Repinique lane with the mutation/fill budget and phrase gating for calls (Rule 4).
6. Feel. A subdivision profile of 1.08 0.94 0.94 1.04 on the 1/16 lanes — the Rule 6 shape itself, repeated across each beat’s four sixteenths — with Swing at 0 on those lanes, because the profile does what the swing was approximating. The 1/8 surdo and the repinique keep a light swing; Rule 6 is a statement about the sixteenths within a beat. Humanize ≤ 0.2.

Rule-Checked Batucada

Lane	Role	Steps	Hits	Rotation	Subdivision	Velocity	Ghost	Mutation	Swing	Timing
1	Surdo pair (2 heavy)	8	2	0	1/8	115	0	0%	0.15	—
2	Tamborim telecoteco	16	7	2	1/16	90	0	0%	0	1.08 0.94 0.94 1.04
3	Caixa carpet	16	14	0	1/16	65	55	5%	0	1.08 0.94 0.94 1.04
4	Agogô	16	5	0	1/16	80	30	5%	0	1.08 0.94 0.94 1.04
5	Repinique (commander)	16	6	3	1/16	95	40	25%	0.20	—

What Breaks the Idiom

• Beat-one weighting. The heaviest low stroke on 1 instead of 2 — instantly a march, not a samba (Rule 1).
• A gappy caixa. Thinning the carpet with density modulation removes the roll (Rule 3).
• Distributed improvisation. Mutation across many lanes; the batucada is ostinato-locked with one commander (Rule 4).
• Polymetric layers. Foreign cycle lengths against the 16-grid (Rule 5) — samba’s evolution is the composite groove, not cycle divergence.
• Deep swing. 0.4+ swing turns the LSSL profile into a shuffle; Brazilian feel is shallower and position-shaped, not pair-shaped (Rule 6).

Sources

Primary: Sandroni 2001 (the telecoteco/Estácio paradigm and its history), Gerischer 2006 and Naveda et al. 2011 (measured microtiming), Fryer 2000 (African heritage and instrument roles). See also refs 42, 43.

Attribution: the telecoteco/Estácio-paradigm analysis (Rule 2) is Sandroni 2001’s; the microtiming profile (Rule 6) restates the measurements of Gerischer 2006 and Naveda et al. 2011; instrument roles (Rules 1, 3) follow Fryer 2000 and standard batucada pedagogy. Rules 4, 5, and 7 are this guide’s distillations of ensemble practice rather than claims stated as rules in one source. All patch values are ours.