The roots of the Harrier lie in the mid-1950's, a time when the possibility of vertical take-off, fixed-wing aircraft had begun to be investigated in several countries. Airframe and engine designers were stimulated by the growing military awareness of the vulnerability of large air bases to tactical nuclear weapons, while the opportunity to operate civil airliners from the centre of cities provided a further impetus. Most importantly, the rapid progress in engine and airframe technology finally made such aircraft appear truly practicable.

There had already been proposals, mainly German and American, for rocket, jet and propeller driven aircraft that took off vertically, being launched from a gantry or rail, or from sitting vertically on a tail-mounted undercarriage. However, although such aircraft had flown, in the shape of the Bachem Ba349 Natter of 1944, the Convair XFY-1, Lockheed XFV-1 and Ryan X-13 of a decade later and the attempts at launching F-100 Super Sabres with rocket boosters (ZELL), none had proven practical to operate in day to day service. What was clearly needed was a propulsion system that would not over-compromise the operational aspects of a VTOL aircraft, while issues of aircraft control peculiar to flying below the stall also needed to be addressed.

Britain, and especially Rolls-Royce, played an important part in such developments. The Derby based firm test flew the twin-Nene powered Thrust Measuring Rig, or 'Flying Bedstead', in 1953, providing them with ground-breaking experience in low-speed and hovering flight. However, the main line of development at Rolls-Royce was not towards single (or twin) engines used for both vertical and forward flight, but rather in the development of specialised lift engines, a concept conceived by the company's Chief Scientist, Dr A.A. Griffith. The rationale behind this scheme was that by using a bank of small lift engines for take-off and landing, each with a high thrust to weight ratio, the airframe and forward propulsion system could be optimised for the cruise condition. This optimisation would hopefully result in vastly improved efficiency, offsetting the parasitic weight of the lift engines. These proposals found their earliest, and most ambitious, expression in Griffith's scheme for a supersonic, vertical take-off airliner to fly from London to Australia, power for take-off and landing being provided by several dozen RB108 lift engines.

The lift engine concept was warmly received by the British Government, leading to the issue of an official specification for an experimental aircraft to underpin the theory. Two companies that tendered to this specification, ER 143T, were Avro and Shorts. The Avro proposal was to modify one of their 707 delta test aircraft with six RB108s in the centre fuselage, able to tilt to provide some forward thrust in lieu of a separate propulsion engine. However, it was Shorts' proposal, the delta wing, fixed undercarriage SC1, with four RB108s for vertical flight and a single RB108 propulsion engine that was successful, the first of two prototypes (XG600) flying conventionally in 1957, tethered hovering tests commencing during the following year.

Although the lift engine concept that underpinned the SC1 was simple in theory, its practical application proved more difficult. In order to be safe to fly the SC1 had to be equipped with a full authority auto-stabiliser, utilising an analogue computer. The failure of a single lift engine would lead to an instant asymmetric lift force, which could only be safely countered by the triplex auto-stabiliser system. While this level of redundancy was designed to ensure that no single failure would cause an accident, in 1963 the second prototype (XG905) crashed when all three auto-stabiliser gyros failed, killing the pilot.

Despite this tragic accident, the SC1 contributed a vast amount of knowledge to Shorts, Rolls-Royce and the government establishments involved, pioneering the investigation of such matters as hot gas ingestion, ground erosion and short take-off techniques. This experience, allied with the investment of money and effort involved, meant that Rolls-Royce became firmly wedded to the lift jet scheme as the most effective solution for vertical take-off jet aircraft, including fighters. The prospect of selling several engines, albeit relatively simple ones, for each aircraft produced offered a tempting vision for the company. However, by the time that the SC1 began flying, a competing proposition had appeared whose origins lay across the English Channel.

French interest in VTOL was stimulated at the same time, and for the same reasons, as British interest. The first concrete results were to come in 1956, when the engine company SNECMA began tests of the Atar powered C.400, a tail-sitting test vehicle reminiscent of Ryan's earlier tests with Rolls-Royce Avon powered rigs that had preceded the X-13. The French experiments culminated in the tests of the C.450 Coleoptere in 1959, like the X-13 a tail-sitting aircraft, albeit one with a radically different annular wing.

Wibault's proposed powerplant (left) for his Gyroptere (right), with four vectoring blower nozzles.

Despite this officially sanctioned effort one French aircraft designer, Michel Wibault, came to regard aircraft that could take-off vertically from a more conventional attitude as a better solution for a VTOL combat aircraft - the Americans having already discovered the severe piloting difficulties of taking-off and landing lying on one's back. Wibault began to sketch the design of a single-seat, single-engine fighter to be used mainly for tactical nuclear strike missions. In order to achieve vertical lift he envisaged the use of four centrifugal blowers disposed around the aircraft's centre of gravity. These would deliver compressed air via rotating nozzles on the blower casings, power being delivered via gearboxes from a Bristol Orion engine, at 8,000 shp the most powerful turboshaft engine available. Residual thrust from the Orion would exhaust at the rear of the aircraft, a cascade of vanes allowing this thrust to be vectored in addition to the main component from the blowers.

Wibault called his design the Ground Attack Gyroptere. Although by no means the first proposal for an aircraft featuring what became known as vectored thrust, the Gyroptere saw the first application of the four nozzle, single engine layout that was to become the hallmark of the Harrier. During 1955/56 Wibault approached both the French and US governments in the hope of gaining support for his project, only to be rebuffed. Despite this, NATO's American funded Mutual Weapons Development Programme, based in Paris, showed some interest in his scheme. The head of aeronautical development at MWDP, Colonel Johnny Driscoll, soon passed on Wibault's March 1956 brochure to Bristol Aero-Engines in England. At the time Bristol were developing the Orpheus engine as part of the MWDP sponsored Fiat G.91 light fighter programme, thereby ensuring that the two organisations already had a good understanding of each other. Bristol's Technical Director, Stanley Hooker, instructed that in the light of this previous contact a serious study of the Wibault proposal should be undertaken.

The main burden of this task fell on the shoulders of Gordon Lewis of the company's Project Office. He soon saw that the poor efficiency of the centrifugal compressors could be overcome by replacing them with an axial-flow fan using the first two low pressure stages from the Olympus 21 engine, driven by an Orion via 1.5:1 ratio gearbox .The airflow from the fan would exhaust via two lateral, vectoring nozzles, with both the core and the fan having separate air-intakes. This engine design, numbered BE.48, was completed by early August 1956 and Wibault sketched a revised Gyroptere around it. However, the BE.48 was quickly replaced by the BE.52 proposal at Bristol. In this, Lewis replaced the Orion with an Orpheus, which was not only lighter, but also allowed the fan, now using three Olympus stages, to be directly driven by a low-pressure turbine added to the Orpheus. This eliminated the gearbox, saving further weight. The BE.52 was the subject of a joint Lewis/Wibault patent taken out in January 1957, which included provision for a pair of rear vectoring nozzles in place of the previous cascade assembly.

With this much-refined engine Bristol felt able to begin soliciting financial support from MWDP, in order to move beyond the design stage. Colonel Driscoll's successor, Colonel Willis 'Bill' Chapman, was firmly behind the proposal, and in mid-1958 agreement was reached for MWDP to fund 75% of the redesignated BE.53, with Bristol funding the other 25%. In the meantime, Bristol had sent brochures out to industry to see just what kind of aircraft could be designed around their radical new engine.