Why must an organisation represent the idea of an individual, rather than the organisation acting in its own capacity while the individual simply sits as its head?
In other words, why do case workers of the ICO form the Information Commissioner's Office, rather than the Information Commissioner simply serving as the head of the Information Commission or Information Rights Office or something like that?
What function does this notion of a corporation sole serve?
The Information Commissioner is the successor of the Data Protection Registrar, created by the Data Protection Act 1984. The office was renamed in 1998 and again in 2000, and the functions have changed along the way. The current governing legislation is the Data Protection Act 2018, schedule 12. There have been many suggestions of changing the structure - such as the ICO's Triennal Review of 2015, and the Leveson Inquiry Report (see sections 4.8 and 4.9, p1109 of volume 3) of 2012 - but that has not happened.
So your sense that the "corporation sole" model is a bit unusual is shared by many distinguished minds, and even in 1984 it was probably anomalous, judging by a quick scan of contemporaneous legislation. For example, the "Cable Authority" established in the same year was created as a statutory corporate body, to regulate cable television. Ultimately, the Data Protection Registrar and the Cable Authority could do the same sorts of things, as far as employing people, acquiring property, entering into contracts, and so forth. There were special rules for how the Registrar and the Authority members would be appointed - these are different rules from one another, but they are both set up by primary legislation. The differences therefore fall to be explained as a matter of policy preference, since they are both legally viable ways to establish a body performing public functions "at arm's length" from ministers.
The main difference is at the leadership level, since as far as doing the job, people employed in the ICO are in the same position regardless of whether they work for a corporation sole or aggregate. If there is one person in charge, then you have a single "head" salary to pay, a single selection process to run to find them, and a single point of accountability to government and to Parliament. If there is a board of directors then they all have to get chosen somehow, and responsibility can be more diffuse. But having several people can lead to better decision-making, and could make the organization appear less biased by having balanced representation.
In fact, prior to the Act of 1984, the creation of a Data Protection Authority (rather than a sole registrar) had been proposed many times, most notably by the Lindop Committee which reported on the matter in 1978. The idea of a registrar was developed in the Government's 1982 paper, Data Protection: The Government's Proposals for Legislation (Cmnd 8539), which led up to the Act. That recommended that there be a single independent official who would be accountable for the work, but that he would be assisted by an expert committee. Additionally, a Data Protection Tribunal would take care of the resulting legal disputes. In the actual legislation, the expert committee was not made statutory, because the government felt that it would be too difficult to define what it ought to do in a way that would stand the test of time (see HL Deb 19 July 1983, vol 443 col 1084 for a Lords committee debate on this topic). The "corporation sole" legal structure gives maximal flexibility to how the Registrar would organize the office over time, while still doing the legal mechanics to give it independent personality.
Why must an organisation represent the idea of an individual, rather than the organisation acting in its own capacity while the individual simply sits as its head? . . .
What function does this notion of a corporation sole serve?
A corporation sole is a way of organizing property ownership for an entity that predates the full development of the idea of an entity as a legal person that was developed primarily in Roman Catholic and Anglican canon law before business corporations came into existence, and is a label that was retroactively applied to this arrangement once the concept of a corporation became widespread.
In a corporation sole, ownership of property and other legal rights and obligations are vested ex officio in whomever holds a particular office. This was a compromise, similar to the compromise involved in the proto-corporate concept of a "trust" that formally vests ownership of property in a human being serving as a trustee that automatically transfers to a successor trustee. A corporation sole vested property ownership in a human being at a time when it wasn't obvious that there could be such as thing as an entity that was a legal person and sole proprietorships were the norm in the business world, while allowing for continuous succession of that property with the organization while keeping organizational assets separate from personal assets of the person holding the office in question.
Another purpose of a corporation sole which has contributed to its ongoing use in contexts like the Roman Catholic church is that each corporation sole has limited liability, so when each geographic region or function of the overall organization led by a different official has title to property vested in a corporation sole with limited liability, liability that, for example, makes only the property owned by the corporation sole of on Archbishop liable for that obligation rather than making the entire church denomination responsible for it. It is functionally equivalent to having many wholly owned limited liability subsidiaries of the larger organization.
Since a corporation sole vests the rights and obligations of this legal person in an office within some organization, however, rather than in the holders of transferrable stock ownership in the organization like an ordinary corporation, it doesn't really make sense for a commercial enterprise, as opposed to a non-profit or a government entity. But the non-transferability and non-alienability of ownership of a corporation sole is a feature rather than a bug within an organization that is contemplated to endure forever.
Indeed, a corporation sole is a close cousin of the civil law concept of an usufruct, which is a right in property that conveys the right to use the property and utilize its profits, but not to transfer or alienate the property. The difference is that an usufruct may be used to benefit the usufruct holder personally, while the use of the property of a corporation sole and the use of the profits and proceeds from the property of a corporation sole may only legitimately be used to further the ends of the office with which the corporation sole is associated.
As noted above, a corporation sole even an even closer cousin to vesting ownership of property in a trustee, solely for the purposes of the trust, which can be conveyed by operation of law when a successor trustee takes office. The only difference is that the limited liability aspect of a corporation sole for which the personal assets of the office holder are not responsible is explicit, absolute, and up front in a corporation sole, while it is only implicit and not quite absolute, in the case of the trustee of a trust.
The use of this relict form of organization for a newly created government office in the U.K. is probably an instance of imitation of the organizational details of some other government office in the U.K.
Some government offices in the U.K. were organized as corporations sole in imitation of Roman Catholic and Anglican corporations sole, in the era of U.K. history before the concept of ordinary corporations and governments as legal persons was well established and widely accepted. These ancient U.K. offices, like the Chancellorship, then continued to have this organization form out of inertia.
As result, the corporation sole form of organization continued to be an option on the menu list of how government division in the U.K. could be organized. Someone creating a new government division in the U.K. then decided to copy the corporation sole organizational model, rather than using another model for some reason.
The reason for choosing a corporation sole option rather than some other organizational option, may have been that the predecessor division of the government office that handled the functions of the newly created government office before it was established as a separate division, was organized in that manner. This would, for example, allow the new government office to almost seamlessly copy the organizational documents of its predecessor agency in a thoughtless way without doing any harm and thus saving time and money in the process of organizing the new agency. But I don't know this to be true for a fact.
