A basement in Dartmouth Park starts from an uncomfortable but useful premise: this is the most demanding thing you can do to the house, and the burden of proof sits with you. A below-ground proposal stands or falls on evidence about ground stability, groundwater, drainage, structure and neighbouring buildings rather than on architectural argument, and the policy Camden applies to it has to be established for your address before anything is drawn. We will not tell you how deep you can go or how far under the garden you can reach, because any such limit belongs to the policy as it stands for your address and authority, and it should be confirmed rather than assumed. What we will do is establish the tests the scheme will actually be measured against before design money is spent, so that the answer arrives at feasibility — when it costs a conversation — rather than at determination, when it costs a scheme.
The hillside is the reason a Dartmouth Park basement is a different problem from a basement on flat ground, and it works in both directions. On a plot where the external ground falls, the house does not sit on one datum: the ground can be well up the wall at one end of the building and away from it at the other, so a single new floor slab beneath the house can be deeply buried under the high side and barely below ground at the low side. That is an opportunity — the low side is where daylight, a door, a courtyard and an escape route become genuinely possible without a deep lightwell — and it is a structural complication, because the box you are building has to retain a different height of ground on each face. Retained height, lateral earth pressure and any surcharge from what stands on the ground above are not uniform around the perimeter, and the design cannot pretend they are.
That asymmetry drives the engineering. A below-ground box on falling ground is loaded unevenly, and the loads have to be resolved into a structure that is propped, permanently and temporarily, rather than merely thick. The permanent props are usually the ground-floor structure above and the base slab below, which means the sequence of construction matters as much as the final arrangement: the wall cannot be asked to carry the full retained load before the elements that restrain it exist. Underpinning is the usual method of getting a new floor beneath a standing house — excavating and casting in short, ordered bays so the building above is never left unsupported — and the bay sequence, the propping and the temporary works are a structural design produced by an engineer, not decisions taken by a contractor on the day. We bring the engineer in at concept, because on this ground the structure is the design.
Water is the risk that never goes away, and it is the one about which we will make no local claim. Nobody can tell you what the ground and the groundwater are doing beneath your specific plot without investigating it, and on falling ground the assumption that water simply drains away downhill is exactly the kind of comfortable inference that a site investigation exists to test. So the order is fixed: a site-specific ground and water investigation first, the waterproofing strategy second. A habitable below-ground room is designed to the dual-protection principle set out in the relevant British Standard — typically integral protection built into the structure combined with an internally drained cavity, and external protection where the ground demands it — with a sump, a pump, a power supply and a maintenance access that someone can actually reach in twenty years' time. Waterproofing is a designed system with a maintenance life, not a product bought at the end.
The neighbours are where basements are genuinely won or lost, and pretending otherwise helps nobody. Excavating below the level of an adjoining owner's foundations engages the Party Wall etc. Act 1996, and in a terrace the underpinning frequently has to support the shared wall — and therefore the neighbour's house — as well as your own. Adjoining owners are entitled to appoint a surveyor, to require a schedule of condition before work begins and to ask for movement monitoring during it, and the resulting award is a legal instrument with its own timetable that a planning consent does nothing to shorten. Construction logistics are a live matter too: excavated spoil has to leave the site, and on sloping residential streets access, vehicle movements, hours of work and neighbour protection are commonly matters an authority will want addressed. We prepare the structural information the party wall process needs early, because unresolved neighbour matters, not the planning decision, are what most often stall a basement.
Light, escape and tenure decide whether the space is worth building at all. A below-ground room that people are meant to live in needs daylight, ventilation, a compliant means of escape and a stair that complies on going, rise and headroom — and on a falling plot the low side of the house is usually where those things can be delivered honestly, through a garden-level opening or a courtyard, rather than forced through a deep front lightwell that the conservation area will scrutinise. Tenure comes before all of it. Where the property is a maisonette or a converted flat, the cellar, the sub-floor void and the ground beneath the building are frequently retained by the freeholder rather than demised to you, so ownership and a Licence to Alter have to be established before a designer is engaged. From there the route is ordered: investigate the ground, confirm the policy and designation that actually apply, resolve the structure and waterproofing, build the planning case on the evidence the authority asks for, then produce the building-regulations and tender package. We do not promise approval. On a basement, we would rather tell you early that the answer is no.
