European wine law excludes the addition of water outright; American wine law permits it up to 35% of the volume; German beer law names water as one of four permitted ingredients; and EU spirits law allows it freely but regulates its quality. Four rules, one liquid — and the difference between them is entirely about when in the process the water arrives.
A 2025 study found PFAS in 95% of the American beers it tested, and traced them to the municipal water the breweries used rather than to anything in the brewing. Spirits are reduced to bottling strength with water too — often a third of the finished bottle by volume — and 2026 is the year the rules on that water started to bite on both sides of the Atlantic.
The same water can be described as 150 mg/L, 15 French degrees, 8.4 German degrees or 8.8 grains per gallon — and be classified "hard" by an American agency and "moderately hard" by a British one on the same day. Hardness is not one measurement but a family of conventions, and a buyer comparing water reports across markets is usually comparing units before comparing water.
Every bottle of whisky below cask strength is the result of one of two engineering choices. Either the spirit and the water met in a vat, were rested, gauged and corrected — or they met in a pipe, metered against each other in real time, with no second chance. The finished liquid can be identical. What the two routes demand of the water is not.
Most of the world's spirits duty is levied on alcohol, not on liquid — so the 420 ml of water in a 700 ml bottle of 40% ABV whisky is the one ingredient the excise system charges nothing for. Much of India works the other way, assessing duty on the bulk litre of finished product, which puts every litre of blending water inside the tax base. Here is the arithmetic on both systems, what it does to bottling strengths and cask-strength pricing, and why an untaxed ingredient is the wrong one to buy on price.
A blending-water specification is a list of limits. A certificate of analysis is a list of claims that those limits were met. Between the two sits a set of standardised analytical methods — conductivity by a three-stage pharmacopoeial test, organic carbon by oxidation, trace ions by ICP-MS and ion chromatography, organisms by culture, and taste by a trained panel — each with its own blind spots. Here is what actually measures what, why two honest certificates can disagree, and the questions that separate an audited supplier from a confident one.