Genre Evolution Map

A map of how the seven BMEM strands emerged, overlapped and transmitted to each other, built from the catalogued formation years, documented personnel movements and stated influence citations. All figures come from analysis/outputs/temporal-summary.json and influence-analysis.json; the lineage edges come from data/networks/bmpn-multilayer.json.

Read this as a map of the catalogue, not of the country. 63 acts are catalogued against a target of 300, and the sample over-represents Dhaka rock and metal. Where the map would mislead if read as a census, it says so.

---

1. Strand lifecycles

Three things stand out.

The strands did not arrive in the order the standard narrative gives. Hip-hop's first catalogued act (1993) predates alternative/indie (1996), folk fusion (1997), progressive (1999) and pop (2002). The genre usually described as the newest arrival is the third-oldest in this catalogue.

Heavy Metal stopped recruiting. Its latest catalogued formation is 2007; alternative/indie kept forming acts until 2018 and hip-hop until 2024. Whether this is a real closure of the metal strand or an artefact of the catalogue having been built outward from a well-documented metal core is not settled by the data. Both readings are live.

The 2000s produced 26 of 62 dated formations — more than the 1970s, 1980s and 1990s combined. That decade is the hinge of the whole map.

2. Emergence and transmission

Diagram source — renders as a diagram on GitHub.
flowchart TD
    FOLK["Folk substrate<br/>Baul · Bhatiali · Bhawaiya<br/>Nazrul Geeti · Rabindra Sangeet"]
    ADH["Adhunik Bangla<br/>+ Liberation-era song"]
    AZAM["Azam Khan / Uccharon<br/>early 1970s"]
    MSR["Mainstream Rock<br/>Souls 1972 · Feedback 1976<br/>Miles · Nagar Baul · LRB 1991"]
    HM["Heavy Metal<br/>Warfaze 1984 · Rockstrata 1985<br/>cover phase → Bangla originals 1991"]
    HH["Hip-Hop / Rap<br/>Ashraf Babu &amp; Charu 1993<br/>Deshi MCs · Uptown Lokolz · T.O.R. 2005"]
    ALT["Alternative / Indie<br/>Shironamhin 1996 · Black 1999<br/>Meghdol 2003 · Arbovirus"]
    PROG["Progressive Rock/Metal<br/>Artcell 1999 · Karnival 2006"]
    FF["Folk Fusion<br/>Lalon Band 1997 · Chirkutt 2002<br/>Arnob · F Minor 2016"]
    POP["Pop<br/>Habib Wahid 2002 · Tahsan<br/>Minar · Imran · Nancy · Kona"]
    PLAT["Platform era<br/>YouTube → Spotify<br/>Coke Studio Bangla 2022"]

    FOLK --> ADH --> AZAM --> MSR
    FOLK -.folk motifs retained across every strand.-> FF
    MSR -->|personnel: Ayub Bachchu, Pilu Khan| MSR
    MSR --> HM
    HM -->|Rockstrata cited by Poizon Green| HM
    HM --> PROG
    HM --> ALT
    MSR --> ALT
    ALT --> POP
    ALT -->|Black → Tahsan| POP
    HH --> HH
    PROG --> PLAT
    ALT --> PLAT
    FF --> PLAT
    HH --> PLAT
    POP --> PLAT
    FF -.Jalali Set × Nigar Sumi.-> HH

    classDef root fill:#efe7d8,stroke:#8C6D31,color:#3a2f1c;
    classDef rock fill:#dceaf5,stroke:#0072B2,color:#0d2c42;
    classDef metal fill:#f7e0d2,stroke:#D55E00,color:#4a2107;
    classDef alt fill:#f5e2ee,stroke:#CC79A7,color:#48213a;
    classDef hh fill:#ddf0fb,stroke:#56B4E9,color:#10394f;
    classDef pop fill:#fbeed2,stroke:#E69F00,color:#4d3500;
    classDef ff fill:#efe7d8,stroke:#8C6D31,color:#3a2f1c;
    classDef prog fill:#d8efe6,stroke:#009E73,color:#0d3c2e;
    class FOLK,ADH root
    class AZAM,MSR rock
    class HM metal
    class ALT alt
    class HH hh
    class POP pop
    class FF ff
    class PROG prog
    class PLAT root

Solid arrows are transmissions with documentary support in the catalogue — either shared personnel, or one act naming another. Dotted arrows are continuities asserted in the literature and the project's own notes but not carried by a structured edge.

3. The six documented domestic transmissions

Only six citations in the catalogue name another catalogued Bangladeshi act as an influence. They are worth listing in full, because they are the entire observational basis for "internal transmission" in this dataset:

Six links across 63 acts is very thin. It reflects how the records were written — global influences were researched systematically, domestic ones only where a source happened to state them — more than it reflects how much the scene borrows internally. Making domestic influence a first-class field in data/schemas/artist.schema.json is the fix, and it is listed in the roadmap.

4. Influence vectors by strand

From analysis/outputs/influence-analysis.json, the most-cited named global influence per strand:

Counts are of distinct citing acts. Token kinds are resolved once across the whole corpus, so an influence cannot rank as a named act in one table and a tradition in another.

The naming pattern is itself the finding. Metal and progressive acts name specific bands. Pop names exactly one act across seven bands, and folk names none at all. Two explanations, and this catalogue cannot separate them: either technical genres transmit through identifiable models while pop and folk transmit through diffuse convention, or metal and progressive musicians are simply interviewed about their influences more often. The second is likely to be at least partly true, since the metal scene has an academic literature (Quader & Redden 2015; Quader 2016) and the pop strand does not.

Ten influences are cited across more than one strand. Metallica (6 acts), Megadeth (5) and Pink Floyd (4) are the scene-wide vectors; Pink Floyd is the only one reaching three different strands.

5. Release-format evolution

From releases_by_decade:

Singles are absent from the catalogue before the 2010s and are 40% of 2020s releases. Against a catalogue-wide median gap of 5 years between releases, this is the album-to-single transition showing up in the discographies themselves. Two cautions: the 2020s are incomplete, and older singles are simply less likely to have been catalogued than older albums. The direction is well supported; the magnitude is not.

6. What the map cannot show

• No sonic measurement. Every claim about how a strand sounds over time rests on documented description, not on audio features.
• Two acts in the progressive strand. Artcell and Karnival. Any statement about "the progressive strand" is a statement about two bands.
• The 2020s look empty (1 formation) because recent acts have not yet accumulated the documentation this catalogue is built from — not because band formation stopped.