by Kelvin Ho
The Spandeck test established in Spandeck Engineering (S) Pte Ltd v Defence Science & Technology Agency1 [Spandeck] laid out a clear framework to determine duty of care in tort of negligence cases, and was further extended to apply to cases previously brought under the separate nominate tort of occupier’s liability in See Toh Siew Kee v Ho Ah Lam Ferrocement (Pte) Ltd2 [STSK]. This article will evaluate the effectiveness of the Spandeck test in resolving the two main issues that plagued the tort of occupier’s liability as highlighted by the Court of Appeal in STSK, namely (1) the often illogical and arbitrary static-dynamic dichotomy, and (2) the invitee-licensee-trespasser trichotomy.
A brief outline of the Spandeck test involves the threshold stage of factual foreseeability, followed by the first stage of legal proximity between the plaintiff and the defendant to find a prima facie duty of care imposed on the latter, and then the second stage of whether there are any negating policy reasons to disaffirm this duty of care. Although it is to be applied incrementally by referring to analogous cases at each stage, the court is allowed to establish new categories of duty where appropriate.
The first problem of the static-dynamic dichotomy regarding the tort of occupier’s liability is that it is never always possible to clearly separate the static and the dynamic. The concept of static refers to the condition of the property while dynamic refers to the activities that take place on it. For example, in the seminal case of Australian Safeway Stores Proprietary Limited v Zaluzna3 [Zaluzna], the plaintiff suffered injuries after slipping on the wet floor of the foyer, which was part of the defendant’s supermarket. The Court opined that on one hand, the condition of the floor was due to the rain and was unrelated to any of the supermarket’s activity, but on the other, the supermarket was running a business at the exact premises where the accident occurred and the very movement of shoppers in the supermarket could definitely be seen as dynamic. As such, the factual matrix in Zalunza seemed to be equally susceptible to both classifications. The use of the Spandeck test can effectively eliminate this issue as it can now be transparently examined under the two stages of the test. With the application of the Spandeck framework, it is no longer necessary to distinguish between static or dynamic. This eliminates the arbitrariness and difficulty of the static-dynamic dichotomy that plagued the tort of occupier’s liability previously.
The second problem addressed by the Spandeck test is the invitee-licensee-trespasser trichotomy, which only arises when the case has been established to be one under the static category. The issue with this trichotomy is that “first (from the viewpoint of logic), it is potentially ambiguous whether an entrant is to be classified as an invitee, a licensee or a trespasser; and second (from the viewpoint of practice), the distinctions between the categories could turn on inconsequential details that potentially lead to injustice”, as opined by V K Rajah JA4. In eliminating the preliminary issue of the static-dynamic dichotomy as explained in the preceding paragraph, the Spandeck test effectively nips this problem in the bud. However, the Spandeck approach is not without its flaws in resolving cases of occupier’s liability.
When the Court first subsumed the tort of occupier’s liability under the general tort of negligence in STSK5, V K Rajah JA emphasized the control of the premises as the key factor in determining a prima facie duty of care in most cases. Admittedly, the concept behind the static-dynamic dichotomy and the invitee-licensee-trespasser trichotomy is intertwined with the idea of control, since there tends to exist a higher degree of control over the static rather than the dynamic, and similarly over an invitee or licensee as opposed to a trespasser. However, the factor of control was not explicitly examined vis-à-vis the Spandeck stages: in particular, whether control should be examined under the proximity stage or the policy stage was left unclear by the Court. Control can be discussed either way: in the proximity stage as the “defendant’s control over the source of the risk of harm to the claimant”, or in the policy stage as the defendant’s control over “a class of individuals of which the claimant is a member”, as noted by Professor David Tan in The Phoenix Rises: Resurrecting Occupier’s Liability Within The Negligence Framework6. Thus, there remains a grey area on whether the factor of control should be examined under the proximity or policy limb of the Spandeck test and this will affect future applications of the test in cases of occupier’s liability.
Assuming control is to be studied in the first stage of proximity, control itself is also insufficient to impose a prima facie duty of care, contrary to what was suggested in STSK that “control of the premises concerned is a sufficient foundation per se for imposing on an occupier a prima facie duty of care under the Spandeck approach”7. Other factors such as the occupier’s knowledge of potential entrants entering onto its controlled premises and the vulnerability of such entrants must be considered in light with the factor of control. For example, in cases where the occupiers are aware of the high possibility that children, who are usually deemed vulnerable, may trespass onto their controlled premises, there should be a closer relationship between the occupier and the trespasser. Thus, it is insufficient to use control as the only factor when considering proximity.
V K Rajah JA made an exception to this by stating that the control factor is alone sufficient when dealing with cases involving lawful entrants8. However, making this reference only seems to bring back remnants of the invitee-licensee-trespasser trichotomy which the Spandeck test initially tried to dismiss, as lawful entrants can be seen as an invitee or a licensee. In fact, the distinction of lawful and residual entrants should no longer be an issue since the enquiry is effectively subsumed under the first stage of proximity in Spandeck; by examining the circumstantial relationship between the occupier and the entrant. Thus, as Sundaresh Menon CJ9 opined, it is better to take full advantage of the flexibility of the Spandeck test itself and do away completely with the distinction between lawful and residual entrants.
In conclusion, the concept of classifying the tort on occupier’s liability under the general tort of negligence, and hence the application of the Spandeck test for cases involving occupier’s liability, was only established recently in the case of STSK in 2013. Even though it was a valiant attempt to make uniform the tort of occupier’s liability via the general Spandeck test, there are still certain unresolved areas as highlighted above, owing perhaps due to the problems in force fitting an age-old tort under an already well-established two-staged test. Nonetheless, the drawbacks of the Spandeck test can still be rectified in future cases, such as clearly defining the factor of control as either a stage one or stage two consideration. Indeed, since the origins of the primary issue of occupier’s liability, being the static-dynamic dichotomy, was “not founded on logic or principle” but rather because “it is rooted in convoluted English legal history” 10, the application of a more recent and established approach such as the Spandeck test in such cases is still very much welcomed.
 Spandeck Engineering (S) Pte Ltd v Defence Science & Technology Agency,  SGCA 37,  4 SLR (R) 100.
 See Toh Siew Kee v Ho Ah Lam Ferrocement,  SGCA 29,  3 SLR 384.
 Australian Safeway Stores Proprietary Limited v Zaluzna,  HCA 7,  162 CLR 479.
 Supra note 2 at .
 Supra note 2 at .
  21 TLJ 59 at page 70.
 Supra note 2 at .
 Ibid at .
 Ibid at .
 Ibid at .
Download the PDF version of this entry here.