Animal welfare prosecutions reported by the Scottish SPCA: 2011-2019

Findings of analysis of animal welfare case data collected by the Scottish Society for the Prevention of Cruelty to Animals.


Key findings

The study included 1,543 charges reported to the Crown Office and Procurator Fiscal Service (PF) by the Scottish SPCA, for animal welfare offences dated 1 January 2011 up until 23 July 2019. There were 873 legal cases involving 1,065 persons. Just under half of these cases (45.7%) involved more than one charge and around a fifth (21.8%) involved more than one person. However, less than 2% of cases involved more than five charges or more than two persons. Overall, the median number of charges and persons involved in a case was one. For cases with more than one charge, the median number of charges was two. For cases where there was more than one person involved the median number of persons involved was two.

Of the 1,065 persons, 30.1% of these had more than one charge from at least one of the cases for which they involved. At least 12 (1.1%) persons were involved in more than one case. Overall, the median number of charges and cases per person was one. For persons with more than one charge, the median number of charges was two. For persons with more than one case the median number of cases was two.

The number of charges rose from 2011 up until 2013, where from 2014 the number declined year on year. The proportion of charges that led to a conviction also varied by year of offence. The was lowest for offences in 2015 (44%) and highest in 2013, where 68.0% of charges led to a conviction.

The reasons for each of these findings require further interrogation of the data and wider policy and practice. Consultation with the Scottish SPCA and Scottish Government policy teams may be able to provide further avenues of investigation in this area. For example, it is possible that we see a peak in 2013 due to increased awareness from campaigns or events, a change in policy or practice, or even a change in staffing at the Scottish SPCA Special Investigations Unit or PF. It may also be useful to consider other aspects of the data itself, such as the types of offence, or how many people or cases were involved by year e.g. where there may have been spate of larger or more serious cases around this time with more people or charges, or even where recording practices may have differed, leading to differences in the data quality itself.

The most common charge type was those covered by Section 19 of the Animal Health and Welfare (Scotland) 2006 Act for unnecessary suffering (59.5%). This was followed by charges under Section 24 (18.5%) for failures to ensure the welfare of animals. Together, these two charge types made up 78% of all charges reported to the PF by the Scottish SPCA with offence dates from 2011 onwards. This is similar to what we might expect given research from Scotland by Battersea Dogs and Cats Home (2017) and the RSPCA (2018) in England and Wales.

The most common offence type was to omit to provide veterinary attention. This represented 29% of all charges. A further 20% of charges were for offences omitting both veterinary attention and adequate nutrition, whilst 18.4% of offences were for failings to meet the needs of an animal. These three most common offence types made up over two-thirds of all charges (68.2%). It is not possible to compare these findings to those from previous research as this is detail not collected by other organisations.

Dogs were specified as at least one of the types of animals involved in 60.3% of all charges. This is similar to what we might expect from previous research (e.g. Arluke and Luke, 1997). Cats were listed in 10.1% of charges which is slightly less than we might have expected based on Arluke and Luke (1997). One possible reason for this could be due to the higher numbers of adolescents observed in the MSPCA files, where Arluke and Luke (1997) also found this age group to be more likely to abuse cats. In addition, many differences in methodology, location and time make these findings difficult to compare directly.

For most offences under the 2006 Act, dogs were the most common animal type involved. However, for Section 29 offences (abandonment) more charges involved cats than dogs. This is a finding that might be interesting to explore further in future research.

Overall, the median number of animals involved in a charge was one and most involved either one (55.9%) or two animals (17.6%). For Section 24 (failing to ensure welfare) and Section 29 (abandonment), offences tended to involve more than one animal, with two animals involved on average.

Just over a fifth (21.7%) of all charges did not lead to proceedings and 11.4% resulted in PF measures. There was a verdict of not guilty for 7.1% of charges, the case was dropped for 2.3% and a small number received a verdict of not proven (0.8%). Over half (56.4%) of all charges resulted in a conviction. This is lower than the conviction rate reported by the RSPCA (2018) of around 92% between 2016 and 2018. However, these figures cannot be directly compared due to methodological differences. As is typical in calculations of conviction rates, the RSPCA figures may only include those proceeded against. They also represent persons rather than charges. Accounting for charges proceeded against only (and excluding charges dealt with using fiscal measures), the conviction rate here rises to 84.4% - although this is still representative of charges only, not persons. This is a similar conviction rate to that reported for persons proceeded against in Scotland overall (86% in 2016-17).[38] The RSPCA do not have the same authority as a reporting agency to the Crown Prosecution Service in England and Wales and prosecutions are pursued privately. This may have some impact on the cases that are taken forward and therefore have influence on resultant conviction rates.

Just under half (45.7%) of all charges were listed as having pled guilty, with 54.3% either listed as having pled not guilty or had no detail for whether this was done so or not. Of those listed as having pled guilty, nearly all of these (97.3%) were in court, with the remaining 2.7% listed as pleading guilty to the PF. This has not been broken down by charge or offence type due to time constraints, but is something that could be considered during future analysis of the data. In addition, no analysis was carried out here on any sentencing discounts that may have been offered as a result of a guilty plea. This discount is not automatically applied when an individual pleads guilty, meaning it might be worth recording whether or not this was taken into account at the time of sentencing.

Many charges resulted in an outcome of more than one disposal type with the most common outcome for charges with a court or fiscal disposal being both a disqualification order and a court fine (19.1%). Under Section 40 of the 2006 Act there is no explicit presumption that a disqualification order be made for these convictions. Yet, there is a requirement for the court to state the reasons for not doing so if this was not done. This is similar to animal welfare legislation applicable to England and Wales, where some have argued that such a stipulation makes it difficult for magistrates to choose not to use such measures (Ryan and Pavey, 2007). This may be in part why of charges with a guilty result in court, nearly two-thirds (64.2%) received a disqualification order. However, no further detail on the specific type of disqualification order, such as the situations in which this apply (owning or working), or type of animals this applies was not included in this data. This might be of interest for some avenues of future research.

Just over a third (34.3%) of charges received a fine in court and 4.6% received a fine from the PF. The average (median) fine amount was £300 overall, £360 for those given in court and £200 for those given by the PF. The average (median) length of a disqualification order was 60 months (five years) and the average (median) custodial sentence was eight months. The latter of these – custodial sentences, were used relatively infrequently (in less than 10% of charges), although this occurred more often for Section 23 offences from the 2006 Act related to dog fighting, where 68.4% of those convicted were given prison time. This is perhaps not unexpected given the seriousness and intentionality of such offences.

Although there were relatively few charges for offences covered by Section 23 (animal fighting), all of those with a guilty verdict in court received a disqualification order. For offences covered by Section 19 (unnecessary suffering) and Section 24 (abandonment), the proportion of those receiving a disqualification order was lower, accounting for 65.6% of Section 19 and 73.7% of Section 24 charges with a guilty outcome in court.

Charges sentenced together – known as ‘in cumulo’ sentencing - limit the amount of analysis that can be done to look at specific penalty amounts by specific types of offence. In addition, numbers in these categories tend to be relatively small once broken down to this level, further limiting what can be achieved. This is not something that can be easily rectified and generally leads to the exclusion of such charges from these types of analysis. Future work could be done to look at these in cumulo sentences and how they might differ from charges overall.

Despite these complications it was possible to establish that of those charges not sentenced in cumulo, the average (median) fine amount for Section 19 offences was £250, just 1.25% of the £20,000 maximum available penalty for this charge. Of the 12 months available prison time, the average (median) time given was five months, representing 41.7% of the available maximum penalty. However, average prison time was calculated using only seven of the 27 charges receiving custodial sentences for Section 19 offences once in cumulo sentences were removed. These seven charges are not likely to be a true representation of all 27 Section 19 charges receiving this disposal. In particular as those excluded were those sentenced in cumulo and therefore involved multiple charges – possibly representing the more serious offences with longer sentences. This, along with the relatively low numbers, may help explain why the median is lower than that previously stated for all charges combined, despite being one of the more serious types of offence.

The average (mean) age on offence date for all charges was 41.3 years, with 6.5% under 21 years. Previously cited research found a similar average age for offences related to animal welfare (Arluke and Luke, 1997; Garrett, 2019), although this did vary by the type of animal and offence in some cases (Arluke and Luke, 1997). This is analysis not carried out here but may be possible in future research - in particular where difference may exist between adult and juvenile offenders. For example, previous research has found that adolescent offenders may be more likely to act as part of a group (Arluke and Luke, 1997). In addition, as this analysis was calculated at the level of charge, not individual persons, future research may also wish to calculate this at the person level.

Postcode details were provided and converted to 2011 datazone so they could be matched with an area based-measure of multiple deprivation (SIMD16), urban/rural classifications and local authority (LA) area. This analysis showed that over a third (37.8%) of all charges were from the 20% most deprived areas in Scotland, compared to just 5.3% from the 20% least deprived areas in Scotland. In addition, over half of all charges were from urban areas (59.1%), with the remaining charges from small towns (13.9%) or rural areas (27.0%). The highest frequency of charges came from South Lanarkshire (9.0%), Fife (8.2%) and Glasgow City (8.0%). As some of the larger LA’s this is perhaps not surprising, although the City of Edinburgh saw relatively fewer charges (3.8%), despite being the second most populated LA in Scotland. More analysis accounting for population size would need to be carried out to look for any significant differences between LA’s. However, within this data it was unclear whether the postcode details provided were those of the address details of the person charged or the locus of offence, that is, where this took place.

Suggestions for future research

Throughout this report, analysis has typically focussed on fairly specific types of offence. However, throughout the research literature related to animal welfare it is claimed that there is a need for more consistent and standardized terminology when discussing animal cruelty and abuse. For example, Munro and colleagues have argued that the adoption of child abuse terminology in cases of animal abuse may help to eliminate confusion and subjectivity and provide a 'common language’ (Munro and Thrusfield, 2001). This would involve using terms such as physical, emotional or sexual abuse and neglect – although it is acknowledged that such terms are not mutually exclusive. For this study the intention was to produce some analysis classifying charges by such types of abuse. However, more work needs to be done to define what offences fall into which categories. For many of these this is not obvious given the detail provided within these data extracts. In addition, some of the most common offences (e.g. omit to provide veterinary attention) could conceivably be classed as physical and emotional abuse and neglect, limiting the usefulness of such measures.

Research surrounding the social circumstances of individuals involved in animal welfare cases is likely to be of interest to many, including social researchers and policy makers. Within this data it was possible to match an area-based measure of multiple deprivation (SIMD) using postcode. However, it is not possible to make inferences about a person’s individual social circumstances from group based measures e.g. not all persons living within a deprived area are themselves deprived (and vice versa). A commonly used indicator of an individual’s social circumstance is occupation. This information could be used to match an individual’s occupation to measures such as the Standard Occupational Classification (SOC), popular in social research. However, it should be noted here that despite the potential for such analysis, the preparation work involved would require a significant investment of time and effort due to the way in which occupation is currently recorded – as an unstandardized, free text field. Any future research interested in using occupation based measures such as this would need to take this into account.

Other demographic characteristics that may be of interest but were not possible to consider here were ethnicity or gender of the accused. These could be regarded as key demographic variables when considering more nuanced examinations of factors related to animal welfare offences. For example, where previous studies have found gender differences by type of animal abuse (e.g. Gerbasi, 2004).

In some cases it is likely that a plea bargain may have been offered where some or all charges may have been dropped after charges were submitted to the PF. The original and adjusted charges could offer the potential for some interesting analysis, such as the types of charges most likely to be accepted or removed and under what circumstances. This type of analysis could be informative for future strategy and decision making. Other details not included in this data which could lead to interesting analysis include details surrounding animal welfare and care notices and appeals (as seen in the RSPCA, 2018 report), the type of court a case was heard in (i.e. Sheriff/Justice of the Peace – each with different limits on the maximum penalties available), and dependant on the status of the Animals and Wildlife (Penalties, Protections and Powers) (Scotland) Bill, whether a case is tried by summary or solemn proceedings.

It may also be interesting to look at any other types of offence committed by individuals accused of animal welfare offences. It could be possible to investigate the potential for data linkage projects using unique identifiers such as criminal record numbers (SCROs) contained within this data. However, the latter of these would involve a significant investment of time and effort and would require advice from data linkage and data protection experts due to the sensitive nature of such a task.

The analysis presented in this report show a selection of some basic descriptive analysis possible with this data. However, taking into account some of the lessons learnt from this project and with less time constraints, more complex analysis using this data would be possible. One example of this might be to use multivariate analysis which considers the influence of multiple factors on an outcome variable of interest, such as considering age, gender and offence type on the likelihood of a conviction or a specific type of penalty or amount.

Some of these more advanced analytical techniques may be best suited to person level analysis, rather than the charge level analysis used throughout the majority of this report. This was in part due to the structure of the data, where this was provided in long (one row per charge), rather than wide (one row per person) format. To carry out analysis at the level of persons, the data would need to be re-structured. However, for an individual with the knowledge to do so, it would be relatively easy to achieve given sufficient timescales.


Further specific recommendations were made directly to Scottish SPCA.



Back to top