Mumu: Papua New Guinea's Earth-Oven Feast and Its Gelatinous Drippings
The mumu (sometimes written as mumu or mumu feast) is one of the most important communal cooking and celebratory traditions of Papua New Guinea and the Pacific Islands. A mumu is essentially an earth oven — a pit is dug in the ground and lined with stones, which are then heated by a wood fire until scorching hot. Once the fire is raked out, layered food is placed directly on the hot stones: sweet potato (kaukau — Papua New Guinea's most important staple crop), taro, yam, banana, corn, and protein (most commonly pork — pigs are central to PNG ceremonial culture and status — as well as chicken, fish, or vegetables). All of the food is wrapped in banana leaves, which seal in the steam and impart a subtle banana leaf aromatics to the food. Soil or more banana leaves are packed on top to create a sealed oven, and the food slow-steams for one to three hours. The mumu equivalent — the umu — is found across Polynesia (Tonga, Samoa, Fiji), Hawaii (imu), and other Pacific cultures. When mumu food or its drippings are brought inside and spill on carpet, the result is a stain with a distinctive characteristic not found in most other food spills: the pork drippings from a mumu are often gel-like or semi-solid at room temperature because of the large amount of collagen (gelatin) released from the pork during the long, slow, sealed steaming. This collagen gel requires a specific first-step treatment before standard enzyme cleaning can proceed.
The Gelatin Problem: Warming Cold Mumu Drippings Before Cleaning
Pork contains collagen — the connective tissue protein found in skin, cartilage, and around muscle fibres. During the long slow steam cooking of a mumu (1–3 hours at high temperature), collagen is hydrolysed (broken down) into gelatin. When hot mumu drippings cool on carpet, the dissolved gelatin sets into a gel — similar to aspic or the jelly that forms under cold roast pork drippings. This creates a unique first-step challenge: if you apply enzyme cleaning solution directly to cold mumu gel on carpet, the solution cannot penetrate the set gelatin evenly, and the enzyme contact with the underlying fat and starch is poor. The correct first step is to loosen the cold mumu gel before any enzyme treatment: place a clean cloth moistened with warm water (50–55°C) over the affected area and leave for 3–5 minutes. The warm water melts the gelatin back into a liquid phase, allowing it to be blotted away and making the underlying carpet fibres accessible for enzyme treatment. Only after this warm soak and blot step should cornflour, lipase, or protease enzyme be applied. Do not use boiling water — temperatures above approximately 60°C may partially set the pork myoglobin protein permanently in the fibres. The gelatin itself is addressed by the protease enzyme step that follows (protease digests the collagen/gelatin protein).
Sweet Potato Kaukau Starch and Banana Leaf Tannins
Sweet potato (kaukau in Tok Pisin, PNG's lingua franca) has a starch content of approximately 14–18% by weight. When steamed in a mumu, sweet potato starch fully gelatinises — it goes from granular raw starch to a thick, sticky, partially transparent gel. On carpet, spilled mumu sweet potato creates a sticky, yellowy-white deposit that adheres to carpet fibres and is not removed by blotting alone. Amylase enzyme (the enzyme that breaks down starch — the same enzyme used for rice, pasta, and potato stains) dissolves gelatinised sweet potato starch. Apply amylase enzyme after the warm-water gel-liquification step and the protease step: (1) warm soak for cold gel; (2) cornflour for surface fat (10 minutes, vacuum); (3) lipase enzyme for pork fat (15 minutes); (4) protease enzyme for pork myoglobin and collagen (20 minutes); (5) amylase enzyme for gelatinised sweet potato starch (15 minutes); (6) hydrogen peroxide for residual myoglobin pigment or banana leaf tannin staining (10 minutes). The banana leaves used to wrap mumu food contribute flavonoids and tannins to the food drippings — these are very pale yellow-green pigments (not as intense as cassava leaf chlorophyll) that respond to hydrogen peroxide in the same way as tea or wine tannins. The peroxide step at the end of the sequence addresses any residual tannin staining from banana leaf contact.
Need Help With a Mumu or Earth-Oven Feast Stain in Sutherland Shire?
We connect Sutherland Shire residents with local carpet cleaning specialists experienced with food stains including collagen-gel pork drippings, sweet potato starch, banana leaf tannins, and Papua New Guinean and Pacific Islander dish spills. Get matched with a local cleaner for your home.
Find a Matched Local SpecialistSutherland Shire Cleaners is a local matching and referral service. We connect you with local providers — we do not supply cleaning services directly.
Papuan Mumu on Carpet: Common Questions
What is the full treatment sequence for mumu drippings on carpet?
(1) Remove solid food pieces with a spoon. (2) If the dripping has cooled to a gel: warm soak — place a cloth moistened with 50–55°C water over the area for 3–5 minutes, then blot (this melts the collagen gel back into a liquid). (3) Cold water blotting (do not use warm water yet for the myoglobin — the warm soak in step 2 was limited to gel-breaking; now switch to cold water for the protein-blot step). (4) Cornflour — 10 minutes, vacuum (for surface pork fat and coconut cream fat if present). (5) Lipase enzyme — 15 minutes, blot (for pork fat). (6) Protease enzyme — 20 minutes, blot (for pork myoglobin, collagen/gelatin, and any protein from the food). (7) Amylase enzyme — 15 minutes, blot (for gelatinised sweet potato kaukau starch). (8) 3% hydrogen peroxide — 10–15 minutes, blot (for residual pork myoglobin pigment and banana leaf tannin staining). (9) Warm water rinse (40°C), blot. (10) Cold water rinse, blot dry. Fan-dry thoroughly. The warm soak in step 2 is the key difference between mumu drippings and most other food spills.
If the mumu spill is still warm and hasn't gelled — do I skip the warm soak step?
Yes — if the mumu dripping is still warm and liquid (was just spilled), skip the warm soak step entirely. Proceed directly from spill removal to cold water blotting. The gelatin in a warm spill is still in its dissolved, liquid state and poses no penetration barrier for the enzymes — proceed normally with cornflour → lipase → protease → amylase → peroxide. The warm soak step is only needed when a mumu spill has cooled and set to a semi-solid gel (typically if the food was carried inside after the feast or the spill was not noticed immediately). In hot weather, gelatin sets more slowly — in Sutherland Shire's warmer months, you may have 30–45 minutes before a mumu dripping gels; in cooler months, it may gel within 10–15 minutes.
Does it matter which sweet potato variety (orange vs purple) was in the mumu?
Yes, in terms of pigmentation — the starch treatment is the same for both: (1) Orange-fleshed sweet potato (most common in PNG): the orange colour comes from beta-carotene (a fat-soluble carotenoid), which is present in the pork fat and sweet potato flesh. The beta-carotene contributes to the orange tinge of the dripping and responds to hydrogen peroxide after fat removal. The peroxide step covers it; (2) Purple sweet potato (less common in PNG, more common in parts of Southeast Asia and the Pacific): the purple colour comes from anthocyanins (water-soluble flavonoid pigments). Anthocyanins are pH-sensitive — they are red in acid and blue-green in alkaline conditions. The correct treatment is: cold water blotting first (to remove the water-soluble anthocyanin from the liquid phase), then the standard enzyme sequence, then if purple tinge remains: apply white vinegar diluted in water (acidic) to shift the anthocyanin back to red before applying hydrogen peroxide. Do not apply baking soda (alkaline) to purple sweet potato staining — it shifts the anthocyanin blue-green and makes it more visible.
The mumu included coconut cream drizzled over the top — does this change the treatment?
Coconut cream adds a fat component (coconut cream is approximately 24% fat, predominantly saturated — lauric acid and myristic acid) and a slightly sweet, aromatic residue that dries to a pale yellowish deposit on carpet. The treatment sequence is unchanged — the lipase enzyme step that addresses pork fat also addresses coconut cream fat (both are triglycerides, and lipase is not specific to animal vs plant fat origin). However, coconut cream fat has a higher proportion of saturated fat than pork fat, and saturated fats can be slightly less efficiently digested by lipase than monounsaturated fats — extend the lipase contact time from 15 to 20 minutes if a significant amount of coconut cream was part of the spill. Coconut cream also creates a distinctive sweet, coconut odour in the carpet after the spill. If any coconut odour remains after the lipase treatment and rinse, apply a baking soda paste (bicarbonate of soda with a little water) to the area, leave for 30 minutes, and vacuum.