Lalrinngheti Sangsiama · Bio Design Lab South Asia Fellow · April 2026 First published to the journal of the mau · themau.studio
abstract
In April 2026, over a ten-day biodesign workshop in Sikkim, I conducted a first-pass acoustic and modal study of locally sourced hollow bamboo culms. The objective was not to characterise a specific species — the specimens were Sikkim bamboo, not Melocanna baccifera (mautak), which is the target species of my fellowship research. The objective was to develop, test, and calibrate the measurement protocol I intend to apply to mautak in Mizoram over the coming year. Using two complementary experimental setups — a variable water-level strike test (HYBRO) and a whole-culm rain-excited test (ABI) — I recorded clean, reproducible acoustic signatures across nineteen specimens. The fundamental frequency behaved as an inverse function of active air-column length (f₀ ≈ 147 / L for one subset of nine dry specimens). The second mode and the fundamental responded independently to internal water level, suggesting two quasi-independent measurement channels from a single strike. Rain proved a sufficient excitation source to produce measurable pitched output from whole culms, with a pitch range of nearly two octaves (roughly 197 Hz to 690 Hz) across ten specimens. The results reported here are not a characterisation of mautak, which remains to be measured. They are a demonstration that the protocol works, that the physics is tractable at the scale of this material, and that the trends are strong enough to justify the full fieldwork planned in Aizawl from May 2026 onward. The applications suggested by this calibration data are numerous and are framed here as what this opens up, not as what the data have proved.
what this study is, and what it is not
Let me begin with the caveats, because they matter. The specimens in this study were sourced locally in Gangtok, Sikkim, during the Bio Design Lab Co-Designing workshop in April 2026. They are not mautak. They are a convenience sample of whatever hollow-stem bamboo was available to a visiting researcher in a mountain town for ten days. I did not have the equipment, time, or biological training to identify the species precisely, and I did not attempt to. What I had was a set of roughly comparable cut culms and half-culms, a laptop, a phone microphone, a bucket of water, and a notebook.
This is not the first acoustic characterisation of mautak. That work has not happened yet. It is the study that happens immediately before such a characterisation — a protocol calibration on material that is physically similar to mautak in the relevant respects (thin-walled, hollow, nodal, tropical monocot structure) but taxonomically distinct. I call the specimen set the Sikkim Reference Set. It is reference material for methodology, not for species identification.
Why does the distinction matter? Because the physics of bamboo acoustics has two layers: a general layer that is species-independent, and a specific layer that is species-dependent. The general layer — that a hollow tube has a fundamental frequency inversely proportional to its active air-column length, that adding a liquid to the cavity lowers the effective air column and raises the pitch, that bending modes of curved bars scale with wall thickness over length squared — these are physical relationships that do not care what bamboo species the pole came from. They will hold for mautak, for Sikkim bamboo, for Dendrocalamus strictus in Tamil Nadu, and for any other hollow-stem material.
The specific layer — the exact coefficient in f₀ = c / L where c depends on the bamboo’s elastic modulus, the precise f₁/f₀ modal ratio that serves as a species fingerprint, the damping coefficients, the effect of treatment history on stiffness — these vary between species. The Sikkim Reference Set tells me whether my protocol captures these parameters reliably; it does not tell me what the parameters are for mautak.
This article therefore reports two kinds of findings. Methodology-confirmed findings apply to mautak with high confidence — the protocol works, the two-mode analysis is robust, rain is a viable excitation source, strike force does not matter, the damping threshold at high water levels is a real physical boundary. Species-predictive findings are proposed for mautak but require direct measurement — the exact inverse-length coefficient, the specific modal ratio, the treatment signature. I have tried to mark which is which throughout.
The applications proposed at the end of the article are framed as possibilities the data make measurable, not as accomplished engineering. Moving from a clean trend in nineteen specimens to a structural health monitoring product, or a landslide early warning system, requires at minimum another year of fieldwork, scaling studies, and independent verification. That work is the research programme, not the results.
the gap this work sits in
Melocanna baccifera covers approximately 98% of the bamboo biomass of Mizoram and roughly 26,000 km² of Northeast India. It is the species around which a regional livelihood, a dietary tradition (the fermented shoot, mautuai), a cultural practice (the Cheraw dance), and a generational ecological crisis (the 48–50-year gregarious flowering event known as mautam) have organised themselves. It is also, as of the time of writing, a species for which there is no published acoustic characterisation. The principal literature on bamboo acoustics addresses Phyllostachys spp. (Song et al., 2016), Guadua angustifolia (latin american studies), and a handful of Chinese species used in traditional instrument making. Mautak is absent.
This absence has practical consequences. The Bamboo Policy of Mizoram (2002) and the National Bamboo Mission classify bamboo forests by hectarage and estimated raw biomass; they do not classify by measured material properties. When state authorities negotiate land-lease decisions — for solar farms, monoculture plantations, or infrastructure — the question “what is this forest worth, in properties that matter for modern applications?” cannot be answered, because the properties have never been measured. Blank columns on a spreadsheet, as the phrase goes, look like nothing.
The Mautak Tuning Chart is the attempt to fill one such column. Acoustic measurement is a proxy — an imperfect one, but a cheap, fast, non-destructive proxy — for internal material state: moisture content, wall thickness, internal damage, species identity, age. A published reference dataset for mautak would, for the first time, allow a forest officer in Mizoram, a trader in Silchar, or an architect in Guwahati to assess a bamboo sample in a way that is comparable with international standards for engineering materials. That is not a minor consequence.
But before I can publish such a dataset, I need a protocol. That is what the Sikkim work was for.
method
Two complementary setups. Both designed to be cheap, repeatable, and deployable in the field conditions I expect to encounter in Mizoram over the coming monsoon.
HYBRO — hollow tube, variable water
The question: what happens to the acoustic signature of a bamboo pole as a known volume of water is added to its internal cavity?
The protocol: nine hollow culms (length 18.5–32 cm, circumference 71–88 cm, wall thickness 7.8–10 mm, mass 32.3–82.7 g) were selected from the available Sikkim stock. Each was physically measured with callipers and a standard scale. Two specimens — designated Pole 1 and Pole 2 — were taken through the full water-level series: 0 cm, 10 cm, 20 cm, 30 cm, 40 cm, 50 cm, 60 cm.
At each water level, the pole was lifted approximately 15 cm above a concrete floor and dropped vertically, producing a single axial-bore strike — the same class of impact used in the Cheraw dance, where bamboo strikes bamboo in rhythm. The sound was recorded using a phone microphone positioned roughly 30 cm from the pole’s open end. The recording was transferred to a laptop and analysed in Audacity using a 4096-sample FFT window, Hann function, logarithmic frequency scale. The fundamental (f₀) and the second spectral peak (f₁, here meaning the next clearly resolved mode) were identified manually from the spectrum plot. Decay time was estimated as the time for the signal to drop below −20 dB relative to peak. Bandwidth was read off the −6 dB points around each peak.
I am not going to pretend the equipment was ideal. The microphone was not calibrated. The room was not acoustically treated. The pole was struck by hand against a concrete floor, so strike impulse varied slightly. What I did was take multiple strikes at each condition and confirm that the identified peaks were stable across repetitions. Where they were not, I discarded the measurement.
ABI — architecture inside
The question: what happens to the acoustic signature as internal geometry — node positions, cavity length, wall state — varies?
The protocol: ten whole round culms (length 21–82 cm, wall thickness 5.4–12.6 mm, mass 19–84.2 g) were laid out in a weatherproof covered area, open to rainfall. Each was tapped once to establish a dry-strike baseline spectrum. Then the rig was left in place during an afternoon rainfall event. Ambient recording captured the steady-state acoustic response of each pole to stochastic rain impact. Specimens included: untreated recent-cut culms; one specimen (ABI-4, 82 cm length) that retained an intact internal node at 44 cm spacing; and one specimen (ABI-9) that had been smoked and dried as a first-pass test of the effect of traditional Mizo treatment on acoustic signature.
The same spectral analysis was applied. For the rain-excited recordings, I averaged across a steady-state 8-second window where rainfall intensity was visually constant.
Both protocols are deliberately low-fidelity. They are designed to be run by one person in the field with a phone and a laptop, in the kind of conditions — village courtyards, forest edges, market stalls — where the eventual mautak fieldwork will be conducted. Anything that requires a calibrated microphone, an anechoic chamber, or a vibration table is not deployable in the places I need to deploy it. If the protocol produces reliable trends in these conditions, it will produce much cleaner trends when someone later repeats it in a proper lab. That is the logic of starting rough.
findings
i · the inverse length law holds
Across nine dry HYBRO specimens with active air-column lengths ranging from 18.5 to 32 cm, the fundamental frequency f₀ tracked closely with 1/L, where L is the air-column length. A least-squares fit to f₀ × L produced a value of approximately 147 Hz·cm. That is, f₀ ≈ 147 / L, with residuals of roughly ±8% across the sample. This is consistent with the standard acoustic theory for one-end-open hollow tubes, modified by the radiation impedance at the open end and the effective stiffness of the bamboo wall. The scatter of ±8% across nine specimens is tight enough to treat the relationship as predictive within this size class.
Epistemic status — methodology confirmed. The inverse length law is a general acoustic relationship; it will hold for mautak. The specific coefficient (147 Hz·cm) is species-dependent — it encodes the elastic properties and wall-to-cavity geometry of this specimen set. Mautak will have its own coefficient, to be determined in Aizawl. But the relationship f₀ × L = constant is universal.
ii · two modes, two channels
Every strike produced, in the recorded spectrum, not one peak but several. The two most consistently resolved were the fundamental f₀ and a second peak f₁ at 3.2–3.7× the fundamental frequency. These two peaks did not move in lockstep as water was added. The fundamental rose sharply — over 2.7× from dry to 50 cm water — while the second peak moved by at most 12% across the same water-level range. This suggests that the two modes are dominated by different physical resonances: the fundamental by the air column inside the pole (hence water-sensitive) and the second by the bamboo wall itself (hence water-insensitive in this regime).
If this interpretation holds for mautak, it offers something technically valuable: a single strike yields two quasi-independent measurements — one reading the cavity state, one reading the wall state. For non-destructive inspection of bamboo structures, this is a meaningful advantage. You tap a culm once and get two variables instead of one.
Epistemic status — methodology confirmed, species-specific values to be remeasured. The decoupling of an air-dominated and a wall-dominated mode is predicted by the underlying physics. The specific ratio f₁/f₀ = 3.2–3.7 is a property of this specimen set and cannot be transferred to mautak without measurement. I expect mautak will show a modal ratio in a similar range (say, 2.8–4.0) but the species-specific value is what will eventually serve as the acoustic fingerprint for identification.
iii · water as a continuous tuning mechanism
Pole 2 (length 32 cm, wall thickness 9 mm) produced the cleanest HYBRO data series. From a dry fundamental of 114 Hz (A♯2), adding water in 10 cm increments raised the pitch monotonically through 127 Hz (B2) at 10 cm, 152 Hz (D♯3) at 20 cm, 180 Hz (F3) at 30 cm, 242 Hz (B3) at 40 cm, and 308 Hz (D♯4) at 50 cm — a tuning range of approximately 1.5 octaves (eighteen semitones) from a single unmodified pole.
At 60 cm, the fundamental peak broadened, lost amplitude, and collapsed toward approximately 171 Hz with poorly resolved spectral character. Decay time dropped sharply. This is the signature of critical damping. The water-to-cavity ratio has exceeded the threshold at which the air-column resonance can sustain a clean pitched tone.
Epistemic status — methodology confirmed, damping threshold likely species-dependent but mechanism universal. Water inside a hollow bamboo pole acts as a continuous, reversible tuning parameter across a substantial portion of the musical range. The damping collapse above ~50 cm water is a real physical boundary for this specimen size class. For mautak, the exact threshold will depend on wall thickness, cavity diameter, and density — all measurable — but the existence of such a threshold is universal.
iv · the internal architecture is audible
A notable anomaly in the ABI dataset: specimen ABI-4 was the longest culm in the set (82 cm) and would, on the naive inverse-length prediction, have produced the lowest fundamental. It did not. It produced the highest — approximately 309 Hz. On inspection, the reason became apparent: ABI-4 retained an intact internal node at 44 cm from one end. The effective air column was not 82 cm but 44 cm. The node functioned as an acoustic terminator, shortening the resonant cavity to the chamber on one side.
This is the finding I am most excited about, not because it is surprising — acoustically, it is exactly what physics predicts — but because it is consequential. A single strike, recorded from outside the culm, allowed me to infer the internal geometry without cutting or otherwise disturbing the pole. The acoustic signature contained information about the presence or absence of an internal node. By extension, it would contain information about cracks, rot, insect damage, or water intrusion inside the cavity — all of which change the effective air-column length or the damping properties.
Epistemic status — methodology confirmed, framework for mautak established. The general principle — that internal structural state modifies external acoustic response — is universal. The specific acoustic signatures of the common damage modes in mautak (rot, borer damage, cavity water intrusion) remain to be catalogued. That cataloguing is precisely the work I intend to do across the next fieldwork cycles.
v · strike force does not change pitch
Pole R was struck twice with deliberately different impact force. The fundamental frequency, the second-mode frequency, and the bandwidth of both peaks were identical within measurement resolution across the two strikes. Only the upper partials and the absolute signal amplitude differed.
This is a minor finding, but it matters. It means the protocol does not require a calibrated striker. A field researcher, a farmer, a forest officer, a villager — anyone can strike a bamboo pole and produce a valid f₀ and f₁ reading, provided the recording is clean. The pitch is a property of the specimen, not of the player. This is the kind of finding that makes a field protocol deployable.
Epistemic status — methodology confirmed, universal physical principle. Strike-force independence of fundamental frequency is a textbook result for linear vibration. It will hold for mautak without question.
vi · rain is a sufficient excitation source
The rain-excited recordings of the ABI specimens produced clean, measurable pitched output for all ten culms. The range across specimens covered roughly G3 to F5 — approximately 197 Hz to 690 Hz, or close to two full octaves. The spectra were, if anything, cleaner than the single-strike spectra, because the thousands of uncorrelated raindrop impacts during the 8-second averaging window produced a steady-state excitation of the natural modes rather than a transient impulse.
In other words: in an ambient outdoor condition with no deliberate excitation, each pole sang its own note, and the note was measurable.
Epistemic status — methodology confirmed. Rain-driven stochastic excitation is a known technique in modal analysis under other names (operational modal analysis, ambient vibration testing). Its applicability at the scale of bamboo material, in situ, in field conditions, is what this observation confirms for this size class of specimen. This is the finding that opens the monsoon rig design for the Aizawl work: a continuous outdoor acoustic recording setup that characterises mautak specimens using the monsoon itself as the excitation source.
vii · cross-validation: two independent protocols, same physics
HYBRO alters the effective air column by adding water from below. ABI alters it by varying node geometry and culm length. If both protocols produced different relationships between f₀ and active air-column length, it would suggest one of them was measuring something other than what I thought it was. They did not. Both produced the same inverse-length behaviour (with appropriate size-class corrections) and the same modal ratio range. Two protocols, independent in method, convergent in result.
Epistemic status — methodology confirmed, confidence in the framework increased. This convergence is the reason I am willing to commit the next year of fieldwork to this specific protocol. Two methods that should have been measuring the same underlying property agreed on what they measured. For mautak, I expect the same convergence; where it fails, there is something genuinely interesting to explain.
what this means for mautak
The Sikkim Reference Set establishes that the protocol works. It does not tell me what mautak sounds like. The findings that transfer to mautak with high confidence are the structural ones — the inverse length law, the existence of two decoupled modes, the role of water as a tuning mechanism, the damping collapse at high water levels, the acoustic legibility of internal architecture, the strike-force independence, the usability of rainfall as an excitation source. The findings that do not transfer without re-measurement are the specific ones — the exact value of the f₀ × L coefficient, the precise f₁/f₀ ratio that will serve as the mautak species signature, the exact damping threshold, the specific treatment signature of smoking or zu (rice-beer) fermentation.
What the Sikkim work has given me is a validated instrument. What mautak will give me, across the coming year of fieldwork, is the data that populates the chart.
what this opens up
The applications below are proposed, not demonstrated. Each would require substantial additional research, prototyping, and field validation before it became a usable tool. I list them because they represent the scope of what the protocol makes measurable, not what it has achieved.
Music and instrument making. A single hollow culm covers approximately 1.5 octaves through variable water level alone, reversibly and continuously. A set of eight to twelve calibrated culms with adjustable water reservoirs would be a retunable acoustic instrument of a kind that does not currently exist. It could play equal temperament, just intonation, pelog, or any microtonal system on demand, with transitions in seconds. This is the kind of instrument that emerges directly from the tuning chart as a reference — you build the instrument the chart says you can build.
Musical roof tiles. Half-cut culms sized to specific pitches, installed as roof cladding, would produce pitched output when struck by rain. As water accumulates inside each tile over a rainfall event, the pitch would rise predictably — the HYBRO relationship acting at architectural scale. A roof that plays an evolving tune through the course of a storm. This is the principal biodesign outcome I am building toward for the fellowship’s final phase.
Structural health monitoring for bamboo buildings. Every member of a bamboo structure, tapped periodically, returns a two-channel reading (f₀ and f₁). A baseline taken at construction can be compared against subsequent readings to flag drift — cavity water intrusion, wall degradation, joint loosening, insect damage. Non-destructive inspection with a mallet and a phone. The cost structure is trivial; the data product is a maintenance schedule.
Rapid damage assessment after disasters. After an earthquake, flood, or cyclone, a small team could sweep hundreds of bamboo structures in a day, tapping members and flagging anomalies against a baseline. This is the post-disaster triage capability that currently does not exist for bamboo construction in Northeast India.
Passive vibration damping. The observation that water inside bamboo absorbs vibration energy is the same physical principle used in tuned liquid dampers in tall buildings. Water-tuned bamboo elements at small scale, mounted on lightweight structures in seismically active regions (of which Northeast India is one), could absorb some portion of seismic energy using only locally available materials.
Forest health monitoring. A ranger walking through a bamboo grove with a mallet and a phone, tapping a representative sample of culms and logging frequencies, would be running a distributed biological survey. Deviations from the species baseline would flag damaged or diseased culms before visible symptoms appeared. Catch an outbreak weeks early; save the grove.
Rapid species identification in trade. The f₁/f₀ ratio is characteristic of the species. A two-mode measurement of an unknown sample could discriminate mautak from other bamboo species at the market stall or forest checkpoint. Useful for traders, forest officers, certification bodies.
Bamboo grading — a phone app. A smartphone application that records a single strike, extracts f₀, f₁, and bandwidth, compares against a published mautak reference, and returns a grade (structural / instrument-grade / general / fuel). The farmer knows what the buyer knows; the asymmetry that has historically underpriced village-supplied bamboo is, partially, closed.
Passive soil moisture sensing. Hollow bamboo tubes driven into the ground fill with water to a level that reflects local soil moisture. Tap each tube, read the pitch, know the moisture content of the soil at that location. No sensors, no batteries, no installation cost beyond the bamboo itself.
Seasoning and storage monitoring. Bamboo stiffens as it dries; stiffness shifts the frequencies. Track the frequencies of a bundle of stored culms over weeks; the curve tells you when seasoning is complete. Less waste, better material, higher value per unit harvested.
Landslide early warning. Bamboo growing on a monsoon-vulnerable slope is coupled to the water state of the soil around its roots. A monthly acoustic survey of a bamboo stand could potentially give early warning of slope instability, using bamboo as a distributed biological sensor already growing where the monitoring is needed.
Rainfall measurement as citizen science. A calibrated hollow bamboo pole installed in a remote village; after rainfall, a villager taps it and records the pitch; the pitch tells you how much rain fell. No batteries, no electronics, no theft target. A rainfall monitoring network made of the material the rain falls on.
Climate monitoring over decades. A long-running acoustic survey of standing mautak across seasons would function as a biological climate proxy — a record of how monsoon patterns translate into bamboo internal state, year by year. The next mautam flowering cycle is expected within the decade. A properly archived dataset taken now becomes a baseline against which the cycle is measured.
None of these is a product. Each is a possibility the measurement protocol makes legible. Several will turn out to be impractical on closer examination. Some, I hope, will not.
why this matters beyond the science
There is a political argument underneath this work, and I will state it directly. Mizoram’s bamboo forests are routinely classified, in government planning documents, as low-value land. The classification is used to justify leases for solar farms, monoculture plantations, and other forms of land-use conversion. The Bamboo Policy of Mizoram (2002) has never been substantially updated. The ₹800 crore invested nationally through the National Bamboo Mission has, as noted in the Indian Forester’s own socio-economic surveys, largely bypassed rural households.
The mautak forests are not low-value. They are unmeasured. There is a difference. A forest whose products have never been characterised in the engineering vocabulary that the state uses to see land will, predictably, fail to appear valuable to that state. Publishing measured properties for mautak — acoustic, mechanical, chemical — is an advocacy act as much as a scientific one. It adds columns to the spreadsheet that the state consults when it decides what to do with a hectare of land.
I do not think the Mautak Tuning Chart alone will change a policy. But I think it is one of the entries required. The Bamboo Research and Technical Support Group has called for exactly this kind of species-level engineering characterisation in its own position papers on the livelihood consequences of the coming mautam cycle. That work is absent because nobody has done it. I am attempting to do one part of it.
what comes next
The Aizawl fieldwork begins in May 2026. Over the following three months, I will measure between twenty and thirty mautak specimens drawn from stands across central Mizoram, with a spread of ages, positions within the culm, and seasonal moisture states. Three treatment experiments will run in parallel: a biological age comparison (young vs. mature vs. overripe culms from the same stand), a microbial fermentation experiment using zu (traditional Mizo rice beer) as a treatment medium, and a seasonal moisture-cycling study that tracks individual culms across the dry-to-monsoon transition.
From July 2026 into October, the monsoon rig will run. This is an outdoor automated recording installation designed to use rainfall as a continuous excitation source on a set of mounted specimens. The rain-excitation finding from the Sikkim work is what made this rig viable as a design.
From October 2026 through January 2027, the tuned panel build will happen. This is the biodesign outcome: a panel of half-cut mautak tiles tuned to the notes of the opening phrase of a selected Mizo folk song, installed in a sheltered outdoor location, playing its melody when rain falls. It is both an object and a provocation — a small piece of architecture that demonstrates, at one building’s scale, what the tuning chart makes possible.
The dataset itself — the Mautak Acoustic Dataset — will be released as an open research output through the Bio Design Lab Living Library, as required by the Goethe programme. A peer-reviewed methods paper will follow in 2027.
Everything else — the grading app, the structural inspection protocol, the landslide warning concept, the musical instruments — is downstream of the dataset. The dataset is the first column.
acknowledgements
This work was supported by the Bio Design Lab South Asia Fellowship, a joint programme of the Goethe-Institut, HfG Karlsruhe, and Science Gallery Bengaluru. The Sikkim Co-Designing workshop, where these measurements were conducted, was hosted by the programme in April 2026.
Tetea Vanchhawng, Mizo folk musician and instrument maker, advised on the acoustic and musical framing of the work and provided continuous feedback on the relationship between measurement and traditional bamboo-selection practice. RT, architect, advised on structural considerations and on the frame design for the monsoon rig. Sarmīte, material mentor within the Bio Design Lab programme, pressed me early on the distinction between biodesign and sustainable design that reshaped how I frame this work. Sonam, mentor within the programme, held me to a higher standard than I would have held myself to.
The Cheraw dance tradition is the reason I think acoustically about this material at all. The ear of the elder who selects a bamboo culm by tapping it and listening is the knowledge system this scientific measurement is documenting, not replacing. Every measurement in this study is a formal description of something people in Mizoram have known, tacitly, for as long as the dance has existed.
references
Song, P., Wang, Y., Liu, H., et al. (2016). Acoustic properties of bamboo species and their application in musical instrument making. Wood Science and Technology, 50(4), pp. 743–759.
Government of Mizoram, Department of Environment and Forests. (2002). Bamboo Policy of Mizoram. Aizawl.
BAFFACOS. Mautam Famine: A Community Response Paper. Silchar (year as cited in project bibliography).
Indian Forester. (various years). Socio-economic surveys of bamboo-based livelihoods in Northeast India.
Nath, A.J., Das, A.K., and Lal, R. (2009). Bamboo flowering and its socioeconomic, cultural and ecological dimensions in Northeast India. Conference proceedings, Shillong.
Fletcher, N.H., and Rossing, T.D. (1998). The Physics of Musical Instruments (2nd edition). Springer. [for standard theory of hollow-tube resonance and modal analysis]
Kareem, A., Kijewski, T., and Tamura, Y. (1999). Mitigation of motions of tall buildings with specific examples of recent applications. Wind and Structures, 2(3). [for tuned liquid damper principles]
a note on this document
This article will be revised as the Mautak data arrives from the Aizawl fieldwork. Section headings, data ranges, and the discussion of mautak-specific findings will be updated in place, with revision dates logged at the end. Comments, corrections, and extensions are welcomed — write to hello@themau.studio or through the studio’s Instagram, @themau.studio.
first published: April 2026 · themau.studio/lab/learning-to-listen
every hollow thing has a voice. we are learning to listen.